Multiple originals of the same theme by a master, deal with workshop practices and often cause dissenting opinions among scholars as to their autography.
Although our investigation is still germinal - our study on multiple originals with tools of analysis - attempts to bring some clarity on classifications and focuses on masters’ pictorial evolving technique version to version, regardless of their number or supposed attributed ranking, enhancing the idiosyncrasies between the versions in a given workshop.
We picked a first experimental field, with the second half of the sixteen century, challenging our tools on three masters, showing potent and distinctive style experiments throughout maturity. The resulting analysis showed that our toolbox was efficient on the aforesaid diverse productions when gathering the tools results. In some cases, the reading produced by the usage of Euclidian geometry was conclusive enough, to risk assessing a significant comment, while on other cases it proved to be incomplete, and requested to be fed with tracings and compass dials.
About forty versions have been analyzed with our toolbox and the reading produced.
The methodology could show relevance when tested on other versions as well, pending on the accuracy of the toolbox which could be increased to match additional reading requirements. Hence, we ventured to tailor our toolbox on Leonardo's three versions of the Virgin of the Rocks, although scholars' opinions on the master's technique were plethoric.
The purpose herewith, is to allow collectors to visualize a pictorial journey through versions of a same theme. Indeed, scholars master the deciphering of scientific reports whereas collectors may stumble on the technical data. Our toolbox could benefit collectors yearning for liable data other than scientific reports and scholars' technical jargon.
As sixteen century ideas on autography differ from ours, it is not redundant to recall that the production of a master often implied the workshop, and that certain areas were commonly executed or finished by other hands in the workshop. Jill Dunkerton of the National Gallery writes “Verrocchio was in the habit of consigning distinct zones of his paintings to members of the workshop, entrusting them with supplying the drawings for the assigned areas, not just the application of paint following one of his own designs”- National Gallery Technical Bulletin, N°32- Leonardo in Verrocchio’s workshop.
Our point is to sharpen our understanding of pictorial shifts, by widening the scope of visual observation non invasive, adjustable and appropriate tools that allow a morphing of each version. The morphing does not consist in transforming the image produced but to compare it with other images produced by our tools.
Our tools single out consistency, variation, cleavage and shift with transitions phases between versions. Idiosyncrasies may persist whilst style and quality differ in the very same version. In the case of Leonardo's third version of the Virgin of the Rocks, the idiosyncrasies include the fluffy hair, the curly locks of hair, the transparency of the chiffon on the Child and on the archangel's arm. Voices have debated with maestria, about the bond between the master and his workshop.
We therefore chose to focus on what revealed our tracings and our compass dials in order to isolate the possible degrees of involvement of the workshop and their variation version to version. Our dials were adjusted so as to produce incremental data with each image produced and allow the toolbox to be each time redesigned: the dials although memory-mapped, welcome new parameters according to the master under study.
The main issue of our toolbox, was to get a grasp of the style consistency (lights correlation with shadows /color modeling predominance over line modeling /mutations of line correlation with the underdrawing) from significant areas of the painting and collect the data, for a global and cohesive reading of the three versions. The mapping revealed variations of treatment - creating gaps between the versions - in the execution, as for instance the drapery of the archangel lacking the -wet drapery effect- often observed on Leonardo's designs, and observed in the sister versions described as a " billowy drapery" by Jill Dunkerton.
Then, the whole data collected with the toolbox was correlated to the scientific analysis conducted by the Brera and our tools enhanced the scale of pictorial shift tracked if any, was to be tracked. However, the observation of the variations was in some areas, clouded by restoration's induced woes, thus impacting on the data collected by the toolbox, which called for a condition state report.
Our conclusion took into account the degree of restoration observed, when compared to the sister versions.
The painting was examined in 2010 in its frame, with artificial light and natural light first, then with wood’s light (UV).
The scientific report , included pictures taken under infra red light revealing the underdrawing and restoration work; No trace of grid was found.
This version neither reflects the original colors, nor the richness and the airiness of the painting at the time it was commissioned. Restoration prevented further damage of the painting whilst restored readable clarity back but has induced woes, notably in some areas compromising the appreciation.
The areas include : the rocky landscape, the foliage, the lower part of the virgin 's robe, and part of the archangel's cloth.
The past transfer of the canvas saved the main layers but also caused loss of the ground layer in some areas which had to be replaced during restoration.
The faces and hands endured minor but numerous restorations qualified as - The face of the archangel Uriel has been recomposed and is cut at the nose level.
The pinpoint restoration skillfully operated by the Brera Museum, contrasts with previous restorations existing damages) made most probably in the 19th century, using bitumen pigment in order to stop the migration of pigments and binders (polymerization) but causing a side effect - crocodile appearance - of the surface highly visible.
Restorations have induced different cracks/wrinkles which are noticeable:
The faces bear a network of very fine lines characteristic of a gentle aging as opposed to other parts of the painting where cracks are thicker and dense. The latter cracks indicating restorations and/or stress induced intervention and possible violent temperature shift. In our version, the - flat and tense aspect- of the canvas is due to the crushing of the pigments and binding carrying dust and dirt during the transfer from wood to canvas.