A New Connoisseurship: Smart ways to detect forgeries

Project Information

ARTDETECT

Grant agreement ID: 101088056

DOI

10.3030/101088056

EC signature date 18 September 2023

Start date 1 January 2024

End date 31 December 2028

Funded under

• European Research Council (ERC)

Total cost

€ 1 999 430,00

EU contribution

€ 1 999 430,00

Coordinated by

UNIVERSITEIT GENT
Belgium

Periodic Reporting for period 1 - ARTDETECT (A New Connoisseurship: Smart ways to detect forgeries)

Reporting period: 2024-01-01 to 2026-06-30

Summary of the context and overall objectives of the project

While scientific methods for analysing cultural heritage have advanced rapidly, research in forgery detection must become more cohesive to address scholarly confusion and information overload. Connoisseurship plays a crucial role in art history. The ERC-funded ARTDETECT project is laying the groundwork for a new form of connoisseurship by combining the latest advances in art history, technical analyses and data science. This project will enable experts to look towards the future. The project aims to develop 'thin-slicing' tools to quickly identify potential forgeries for further analysis and to revolutionise connoisseurship through a new approach to assessing and interpreting paintings.

Although scientific methods to analyze cultural heritage are being developed rapidly, research in the field of forgery detection (art created with the intent to deceive) is too fragmentary and discipline-bound to face the current scholarly confusion and information overload. Art historians are used to focus on the past and the present. This project will allow art experts –for the first time– to face forward and predict high risks. The goal is to first generate fundamental new insights into significant anomalies that betray forged paintings (as well as into key characteristics of comparable originals); secondly, to develop ‘thin-slicing’ tools, i.e. computer programs and checklists that help to quickly and effectively select potential forgeries for in-depth analysis; and ultimately, to give connoisseurship a new reach and relevance through the novel, integrated approach to assess, categorize and interpret paintings.
Despite the great importance of connoisseurship for art history (providing the basic classification of who created what and when, information essential for virtually all interpretative studies in this field), its methodology has not kept abreast with innovative multi-disciplinary approaches to collaboration among experts in different fields of inquiry. The 21st century has witnessed major breakthroughs in thinking about authenticity within art history; the scientific methods and technical means to examine artworks have improved tremendously; and data science and data visualization techniques allow us to think on an unprecedented scale.

Work performed from the beginning of the project to the end of the period covered by the report and main results achieved so far

-We have been awarded an ERC Proof of Concept grant for the project: COunterfeit DEtection: Validating the ARTDETECT approach with Stakeholders in the loop, a collaboration with the Dutch police, and a concept developed with support from GRACE (Ghent Research institute for Art & cultural heritage Crime and law Enforcement) and the Belgian police. It enables us to document a large amount of (suspected) forgeries intercepted by the police, and create an annotated database (useful to validate insights from the larger research project ARTDETECT).
-At the request of Europol and the Bulgarian police, an extensive expert report (with a clear conclusion) was written on a suspected forgery for an upcoming lawsuit in Bulgaria, using a broad, integrated approach and a range of analytical techniques (currently classified pending the lawsuit).
-ARTDETECT gained access to a wide range of case studies and reference works, and signed agreements which allow us to not only research the works with various techniques but also publish the results.
-Our team built an innovative type of MA-XRF scanner: the ARTDETECTor (with software) which is more affordable and lighter, but also more precise than what is currently on the market.
-Presentations at international conferences and publications, amongst others:
International Atomic Energy Agency meeting at the United Nations in Vienna: IAEA Technical Meeting on Expanding the Stakeholder Base of Nuclear Techniques for Forensic Science: Novel Applications and Niche Areas 30 Sept- 4 October 2024. Keynote lecture on the ARTDETECT project and its aims, and chaired parts of the conference (November 2024).
Presentation of the ARTDETECT project at the Community for European Research and Innovation for Security (CERIS), European Commission, Workshop: Innovation against Trafficking of Cultural Goods in Brussels (December 2024)
Presentation at the Renaissance Society of America (RSA) conference in Boston (March 2025)
Presentation ARTDETECT and CODES at the Belgian Abstract Modernism in the spotlight (BeAM) workshop at the FeliXart & Eco museum, Brussels (December 2025)
Draft article for a special issue of Technical Art History and Heritage Science on a key case study and the goal of the ARTDETECT project.
Small book in Dutch and English on a case study:Twee Vissersjongens: over authenticiteit en vernieuwing in het oeuvre van Frans Hals (1582/83-1666) and Two Fisherboys: On Authenticity and Innovation in the Work of Frans Hals (1582/83-1666)

Progress beyond the state of the art and expected potential impact (including the socio-economic impact and the wider societal implications of the project so far)

Breakthroughs or advances beyond the state of the art are mostly expected in the second stage of the ARTDETECT project (years 3-5) since it is the systematic combination of approaches and the creation of tools and checklists and, ultimately, a new type of connoisseurship that are the most innovative elements. We are currently laying the ground for the innovations by carefully analysing a wide range of case studies with state-of-the art research methods and techniques, as planned in the Description of Action. Nevertheless, some of the methods and techniques we have developed and used are very innovative in their own right. Notably, we custom-designed and built a new, affordable, precise and relatively light type of MA-XRF scanner, which could be developed into a marketable innovation in the future.
Also, we used two ARTDETECT case studies (forgeries) to test and further develop a brand-new technique: MA-Raman scanning, designed by Ghent University's chemistry department. Contrary to existing Raman analyses, this technique allows for the visualisation of entire surfaces (not just point measurements) comparable to MA-XRF scanning. A major challenge lays in the penetration of the varnish layer, which can strongly interfere with Raman measurements.
As per the goals of the ARTDETECT project, the first three projects aim to generate new insights into what particular characteristics distinguish forged paintings from related reference works. Thus far, each case study has indeed yielded significant new insights into the specific materials, techniques, signs of aging and contextual factors that set forgeries apart from the reference works. We suspect that signs of aging are particularly revealing; yet, it is still too early to draw firm conclusions about larger patterns.
As the ARTDETECT project is the first large-scale systematic research project to focus on forgery detection, and we have a much larger body of works at our disposal than originally expected (in part thanks to the new Proof of Concept project CODES), as well as a newly expanded, well-equipped research lab (also unique in the field of art history), we expect our knowledge to increase rapidly over the coming years and enable us to distinguish larger patterns and clues that will play a key role in the future tool development as well as in the theoretical and methodological reflection on connoisseurship.