Science’s war on art fraud 

In 2016 a team of scientists led by David Thurrowgood of the National Gallery of Victoria took a painting by French impressionist Edgar Degas to the Australian Synchrotron in order to solve a long-standing mystery.

Art experts had previously noted that the artwork, Portrait de Femme (1876-1880) had been painted directly over a previous composition. Faint traces of the earlier work were visible but the piece was otherwise completely obscured – probably as the artist intended.

Thurrowgood and the team at the Synchrotron, in the Melbourne suburb of Clayton, used high-definition X-ray fluorescence (XRF) to penetrate the surface of the painting to reveal (upside down, as it were) the face of an entirely different sitter. 

With false colour added to provide at least figurative flesh to the hidden portrait, the result was extraordinary – and a powerful demonstration of how cutting edge science and technology have an increasingly valuable role to play in revealing the secrets of art.   

Nowhere is this more the case than in the murky but highly profitable area of forgery. The global art market turns over something north of US$60 billion a year, and some experts estimate that as much as 50% of the works traded are forged.

Now, however, new techniques are being developed in laboratories around the world that look set to make the forgers’ lives much more difficult.

170426 art fraud full
Penetrating insights: high-tech methods such as X-ray fluorescence are revolutionising art conservation and authentication.
David Thurrowgood

Gumshoe detection

In 2010, German painter Wolfgang Beltracchi was unmasked as one of the most successful art forgers of 

the modern era, reaping millions of euros through creating near-perfect artworks, mainly in the styles of 20th century masters.

His output included works ostensibly by the great Cubist painter Georges Braque (1882-1963). Should anyone today attempt to repeat that dishonest little trick – and someone, inevitably, will – he or she will find attempts to pass off a moody Braque very, very much more difficult.

In 2016, Clara Granzotto and Kenneth Sutherland from the Art Institute of Chicago developed a new imaging technique to investigate the media used by the French artist in creating a painting titled Ajax (1949-54). The work was owned by the institute and catalogued as “oil on paper” but the researchers had a hunch the description was inaccurate.

The pair developed a method of analysing minute particles taken from the edges of the work. Called matrix assisted laser desorption ionisation time-of-flight mass spectrometry (MALDI-TOF MS), the technique uses lasers to ionise large molecules, such as carbohydrates.

When the results were in, Granzotto and Sutherland found the paint mixture contained two separate types of acacia gum. Known in art circles as gum arabic, the substance was a common addition to watercolour paints during the period. This indicated that Braque had used watercolour as well as oil paints to make the piece.

MALDI spectrometry is today used mainly to provide detailed information for conservators and restorers. Should a previously unknown Braque from the same period suddenly come onto the market, however, it’s London to a brick any decent dealer will be giving the gumshoe detectives a call.

Looks deceive no more

Many methods used to determine the authenticity of paintings – scanning electron microscopy, for instance – necessarily destroy part of the artwork itself.

Perhaps the best known non-destructive investigative method is optical coherence tomography, a medical imaging system that uses near-infrared light and is employed often by ophthamologists to get three-dimensional, highly detailed images of the retina.

In the art world it is extremely useful for providing in-depth data on elements such as the composition and layering of paint. Its main drawback, however, is that it images only very small areas, so using it to map a large canvas is both time-consuming and expensive.

Recognising this problem, a team of computer scientists and art historians from the Pusan National University in South Korea set about designing an alternative. Led by Seonhee Hwang, the group developed a method that combined fibre optics reflectance spectroscopy with a laser-based topographic analysis. The system is able to scan an entire artwork, measuring the colour characteristics of the whole piece. At the same time, a laser-based map of the thousands minuscule ridges created by the artist’s brushstrokes and fingerprints is also produced.

To test the accuracy of their new technique, Hwang and colleagues commissioned expert painters to create forgeries of paintings by well-known Korean artists. The system was then used on the originals and the fakes. Writing in PLOS ONE in February this year, the researchers reported that the reflectance spectroscopy identified the forgery in 76% of cases, while the laser topography was successful every time.

It’s in the DNA

Once upon a time an artist’s signature – down there, in the corner of the painting – was just about all the verification anyone needed to be sure an artwork was genuine. If you had the provenance as well – the documented history of the work’s sales and owners – no more proof was needed.

Such innocent days are long gone. Signatures and records of sale can both be forged; and even the experts, from time to time, are fooled. {%recommended 3792%}

Is there, then, a foolproof way to establish that a work is genuine? For new paintings, the answer is yes, and it involves synthetic DNA. A technique developed at the Global Centre for Innovation at the State University of New York involves inserting a tiny amount of specially created genetic code into still-wet paint – establishing a permanent, updatable record a little like a microchip inside a pet cat.

The system was developed at the behest of a company called the ARIS Title Insurance Corporation, which specialises in insuring fine art. Although still in its infancy, the company intends to log the DNA – each piece unique and created to order – into a database, which will also contain provenance information. To verify the authenticity of a tagged work, all any dealer will have to do is run a proprietary scanner over the canvas. 

The DNA bonds with the media used to make the artwork, so it is impossible to remove it, let alone copy the work. The system – dubbed the i2M Standard – is now being trialled, with a full-scale rollout expected soon.

Doing your block

If master forgers often get away with creating fake oil paintings, imagine what they can get away with digitally made art, a medium that can be copied any number of times without the slightest change occurring.

Everyone knows digital art is endlessly reproducible, but over the past few years artists who work specifically in digital media have started to attract big prices for their creations. Since then, two questions have become urgent: how do the artists protect their originals; and how can buyers be sure they are getting the genuine article?

The answer is a blockchain – the same type of recording technology now commonplace in the world of online currencies such as Bitcoin. 

A blockchain is a growing database of individual transaction records (known as blocks). Each transaction produces a timestamp and a link to the previous one – creating a verifiable and (theoretically, at least) forger-proof provenance.

Several companies in the art world are already offering blockchain verification services to artists keen to maintain control over their creations.

In the world of digital art this is quickly emerging as a critical course. In a field where 10 people can display artworks that to all intents and purposes are exactly the same, there has to be some way to verify who has the “real” – and hence really valuable – one.

Please login to favourite this article.