The outstanding exhibition, Rubens, Drawing on Italy, shown in Edinburgh and Nottingham, shed considerable light on the highly complex subject of Rubens’s reworking and ‘improvement’ of drawings by other artists. This highly rewarding – and (at least in Nottingham, where I saw it) surprisingly beautiful – exhibition gave the opportunity to see at first hand the full range of the works that document Rubens’s long-lasting and profound involvement with the arts of Italy. Understandably, the show raised a variety of complex issues, as did the associated symposium, held in Nottingham (November 29-30, 2002); the symposium program included papers, mostly excellent, from Donatella Sparti, Bert Meijer, Nico van Hout, Kristin Belkin, Anne-Marie Logan, Justus Müller Hofstede, Jeremy Wood, Zirka Filipczak and Elizabeth McGrath.

The exhibition contained copies by Rubens himself after Italian works, copies and drawings by other artists retouched or adapted by Rubens, and also drawings by Italian masters which were in his collection but which he did not rework. The copies by Rubens himself included paintings as well as drawings, and indeed one of the highlights of the exhibition’s Edinburgh incarnation must have been the juxtaposition of Titian’s great Diana and Callisto (Sutherland Collection) with Rubens’s full-size painted copy (The Earl of Derby). But as the exhibition title makes clear, it was primarily through the medium of drawing that Rubens looked at Italian art, and one of the exhibition’s main achievements lay in its success in elucidating, through a fascinating stylistic and qualitative spectrum of the works, exactly how Rubens studied Italian art through drawings.

In Nottingham, the drawings were grouped according to the Italian artists to which they related (‘Raphael,’ ‘Michelangelo,’ etc.), so the viewer was able to gain an immediate impression of what aspects of each of these masters’ works particularly interested Rubens, how he reacted to them in his own copies and his reworkings of copies by others, and in some cases which drawings by those Italian artists he owned. In the fine selection of works, and in the text of the catalogue, we see the fruition of Jeremy Wood’s extensive research on Rubens as a drawings collector and on Rubens’s retouching of the drawings in his possession. The catalogue provides the first reasoned account of how Rubens’s copying of Italian prototypes, and his retouching and reworking of copies made by others, evolved during the course of his career. The details of the chronology that Jeremy Wood has set out will undoubtedly be the subject of continuing discussion, as will his dating of certain individual copies/retouchings, but his establishment of a general chronological structure within Rubens’s works of this type is nonetheless a very valuable contribution.

Rubens brought back from Italy a great number of drawings after Italian prototypes, to which he referred for the rest of his life. He made elaborate drawn copies after Leonardo, Raphael, Michelangelo and others (though usually of individual figures or figure groups, rather than whole compositions); he made, and reworked, counterproofs of his own copies (e.g. cats.12-13); he took fragmentary or damaged drawings by other artists and ‘repaired’ or extended them; and he acquired or commissioned drawn copies of important compositions, which he then reworked, retouched and in some cases cut up and rearranged entirely.

Particularly amongst the drawings in the last two of these four categories, Rubens’s intervention was often so extensive that it can be extremely difficult to be absolutely sure where the work of one hand ends and that of another begins or, indeed, whether there really are different hands present in the same sheet. On a certain level, one could argue that it actually doesn’t matter that much whether a particular copy after, say, Polidoro is entirely by Rubens, or is a copy by another anonymous hand which Rubens acquired and reworked extensively, since the primary significance of that drawing is as a record of what aspect of Polidoro’s work or what figures from his compositions Rubens valued enough to want to record and keep, and the end result is always more a reflection of Rubens’s vision of these figures than of Polidoro’s. But on another level the connoisseurship questions do, of course, matter enormously, as the understanding of Rubens’s working method and drawing technique presented in Jeremy Wood’s exhibition catalogue will also define his forthcoming Corpus Rubenianum volume on the retouched copies of Italian works, which will be the standard reference work on the subject for the coming generation. It is therefore a source of some concern not only that quite a number of Wood’s conclusions regarding attribution were hotly disputed by many of the scholars who gathered in Nottingham in November, but also that there was precious little consensus within those dissenting views.

For my part, while in many cases I agree entirely with Wood’s judgement, I do believe that a significant number of the drawings that were exhibited as anonymous copies or original works by other artists, retouched by Rubens, were in fact entirely drawn by Rubens himself. In some of these cases Rubens appears to have made a conscious effort to imitate not only the composition but also the style of an earlier master: most striking perhaps is the British Museum’s drawing after two figures from Bellini’s Frari Triptych (cat.3), in which I simply cannot discern two hands. Similarly unsustainable, in my view, is the assertion that the Fitzwilliam study of Two Robed Men (cat.61, as ‘Attributed to Bartolomeo Passarotti, retouched by Rubens’) is by two different hands. The penwork does indeed bear some resemblance to Passarotti’s style, but without the washes and heightening, which Wood says are later reworkings by Rubens, the figure simply does not exist: the elements of the different media are entirely complementary in this drawing, which has to be entirely by one hand – that of Rubens. I also believe this is true of a number of other drawings, most strikingly the two British Museum drawings, after the figure from Raphael’s Fire in the Borgo (cat.22) and Polidoro’s Man Leading a Horse (cat.33), and the Louvre copy after Raphael’s Virgin and Child with Saint Elizabeth and the Infant Saint John the Baptist (cat.25), the last of which I see as totally comparable in technique and handling to the British Museum’s Pan and Syrinx, after Raphael or Giulio (cat.26), here exhibited, in my view correctly, as a straight copy by Rubens. I also think cats. 16 and 27 are entirely by Rubens, and possibly also cat.15 (definitely by one hand, but perhaps Rubens studio, rather than Rubens himself). Cat.38, though not by Rubens, is surely also by one hand, probably Quellinus.

As for the many drawings in which two stages of execution are indisputably visible, I propose that several of the rather routine underlying copies, which we all agree were subsequently enlivened and worked up by Rubens, are matter-of-fact record drawings that he himself made on arrival in Italy, and returned to at a later date. He made a number of rather uninspiring record drawings of this type – e.g. the series of copies after Michelangelo’s Sistine Prophets (see cats. 9-10) – as well as many more accomplished, freely interpretative copies, often over the base of a routine copy; logically, I would have thought Rubens was no less likely to have used his own copies as his point of departure for the second, freer category of drawing, as he was to have used one of the many anonymous copies that we know he had in his possession. Wood accepts something approaching this creative history for one drawing in the exhibition, the Leonardo-inspired Fight for the Standard, in the British Museum (cat. 5), and to some extent also for the reworked counterproof of Rubens’s own copy after another Michelangelo figure (cat. 13), but I would, on stylistic grounds, extend this to a number of others, including cats. 17, 23, 40 and 68, as well as cat. 36 (rather confusingly, the artist line in the catalogue lists this drawing as entirely by Rubens, but the subsequent entry describes it as a reworking by Rubens of an anonymous, earlier copy).

These opinions regarding attribution also have implications regarding the question of the extent to which Rubens was prepared to interfere with and ‘improve’ the works of other artists. As Wood and others have noted, Rubens does seem to have had a very well developed idea of what was or was not acceptable to rework. I believe this tendency to respect original works by good artists was actually much stronger than Wood suggests, and that Rubens hardly ever interfered significantly with the original drawings by major artists that he owned; the fine studies by Annibale Carracci (cats.71 and 73) and Baldassare Peruzzi (cat.18) are cases in point. As argued above, I do not believe the (partial) attributions to Passarotti, Polidoro and Federico Zuccaro of cats. 15, 16, 27, 61 and 68. Nor do I think the underlying drawings, reworked by Rubens, in cats. 44 and 58 are actually by Taddeo Zuccaro or Battista Franco; the former I think is merely a studio or later copy, the latter a Giulio Romano studio copy after the associated print.

There were some examples in the exhibition of drawings by important artists that also incorporated a significant contribution from Rubens, but almost all of these were, however, ‘repair jobs,’ like the Liverpool Federico Zuccaro drawing of The Conversion of Mary Magdalene (cat. 70), in which Rubens replaced entirely the lost lower left corner of the sheet but did not interfere with the undamaged parts of the composition, suggesting that his respect for Federico’s draughtsmanship may have outweighed considerations of stylistic consistency. Slightly more complicated but still, I would say, consistent with the patterns of intervention or non-intervention that I have described above is the Louvre drawing of A Woman with two Children in Front of a Fire (cat. 75), exhibited as Annibale Carracci, retouched by Rubens. Here, Rubens has taken a small scrap by an Italian artist and has enlarged the composition considerably on all sides, adding further figures which extend well onto the original sheet. In fact, so dominant are the elements introduced by Rubens that it is hard to come to a clear conclusion regarding the attribution of the central figure group; Wood’s assertion that it is by Annibale is not wholly convincing, but even if he is right, this original was such a fragmentary scrap that it is perfectly understandable that Rubens should have had no qualms about revising it extensively.

Of all the drawings in the Nottingham showing of the exhibition, the only one in which I felt a genuinely attributable drawing by an artist of some significance had really been fundamentally reworked by Rubens was the British Museum Assumption of the Virgin, originally by Domenico Campagnola (cat. 56). To this one might perhaps add cat. 43, but Rubens’s intervention in that drawing seems to me minimal, and I am also not entirely convinced of the attribution to Taddeo Zuccaro of the original.

Given the complexities of all these connoisseurship issues, this exhibition could very easily have been impenetrable to all but a select band of cognoscenti. But irrespective of one’s views regarding the attribution of individual sheets, the exhibition presented on several different levels an intriguing and beautiful account of Rubens’s intense involvement with Italian art, and has made a very valuable contribution to our understanding of this rich and fascinating subject.

Gregory Rubinstein

Sotheby’s, London