The Provenance of Artists' Materials
Provenance research focuses on an artwork’s history from the moment it leaves the artist’s studio through to the present day. Scientific analysis extends this timeline further. By enabling us to peer beneath the surface of a work we can view it as a product of its constituent material parts: the support, pigments and binder. Each component has its own provenance, which offers intriguing insight into the artist’s practice and the painting’s geographic origin. The documented history of a painting’s successive ownership after it leaves the artist’s studio is vital in ensuring its legitimacy; in much the same way, the undocumented history of the materials from which a painting is made is important and compelling in revealing its integrity.
Although dendrochronology is often associated with the absolute dating of panel paintings, it can in many instances establish the geographical origin of a panel. Simply put, the width between each concentric tree ring corresponds to a year’s growth. Adverse growing conditions result in stunted growth producing narrower rings, while favourable conditions produce wider rings. When the number of rings and their widths are plotted on a graph, the age and origin of a panel can be calculated. Since trees growing in the same climatic region generally share the same ring patterns, because of the impact of the same stimuli, such as temperature and rainfall, we can establish geographic signatures and identify their presence in panels.
A large number of panel paintings have been analysed using dendrochronology, yielding a sizable database where correlations and patterns become apparent. For example, many western and northern European artists did not use panels produced from trees indigenous to the area in which they lived, rather they purchased imported panels from the Baltics. There are even instances where multiple panel paintings produced in the same workshop originated from the same tree.
Dendrochronology is already used in authentication as it provides a precise and absolute dating technique, yet it also shines a light on artistic practice: if an artist’s workshop only used panels imported from Poland, there may be grounds for suspicion if an unknown painting alleged to be by the same studio was painted on a panel from a different region, such as the Italy, regardless of whether it dated from the correct period.
Lead white, the ubiquitous pigment found in almost all Western art can be traced to its geographical origins. Lead has four naturally occurring isotopes, 204Pb, 206Pb, 207Pb and 208Pb, found in different concentrations in rocks around the world. Accurate tracing is made possible because different regions have different “signature” geochemical conditions, which result in the characteristic abundance ratios among the four isotopes in lead ore. This makes it possible to trace the paint pigment’s lead back to the mine where the ore was extracted.
Although some mines have overlapping isotopic compositions, the variance between some regions are statistically significant. For example, it is possible to differentiate between lead white from transalpine and cisalpine sources. In the case of Van Dyck, an artist who painted on both sides of the Alps, we are able to discern whether the picture was completed using pigment purchased from a Flemish or Italian source, by looking at the isotope abundance ratios alone, which could be utilised to contextually date or authenticate a work.
As is the case with analysing wooden panels, a complicating factor in lead isotope analysis is the international trade of goods and artist’s materials. For instance, despite never visiting the UK, the likely source of the lead used in Vermeer’s Girl with a Pearl Earring is a mine in the Peak District. Given artists’ predilection to use the same colourman however, narrow clusters for the lead white used by individual artists and studios can be identified.
As a painting ages, a fine network of cracks appears on its surface. The artist’s technique and materials, as well as the painting’s environmental history develop its craquelure patterns. These patterns can be analysed mathematically so that correlations within an artist’s body of work are identifiable, or, more broadly, to identify art from a particular region or date. A basic example is to compare the difference between the characteristic craquelure patterns of Italian, Flemish, Dutch and French artists:
The pioneers of this research described craquelure as an unintentional ‘signature’ that develops as a painting responds to its environment. Aside from its use in identification techniques, recreating the craquelure patterns produced by the environment over the time since the work’s production date is significantly challenging. At Hephaestus Analytical we are taking things a step further and are using machine learning to reveal deeper craquelure “signatures” in order to identify characteristic patterns belonging to each artist. Craquelure is one part of a suite of tests that we are building to eliminate forgery from the art markets, and create the highest evidentiary standards in the authentication of art.