Michael Savage
posted 17 January 2006

This exhibitions brings together just six thirteenth and fourteenth century Italian panels, each panel a part of a larger work of art that was separated in the past. Briefly, we are able to see these works as they should be seen. The scandal is the brevity; these loans should not be returned.

Four of these panels have been acquired recently by London's National Gallery, and in one happy case a pair of paintings (a diptych) has been permanently reunited, after the two halves were bought separately in 1999. This is the earliest exhibit - an Umbrian work from around 1260. The relationship between the Madonna and Child on the left and The Man of Sorrows on the right is tender, subtle and profound. The two halves relate perfectly as was intended, with a pathos remarkable for the date. Although this is from just before the conventional 'start' of the Renaissance, the sense of internal feeling is already beyond Byzantine or Romanesque prototypes.

The most exciting thing here is the Cimabue, which is an extraordinary rarity painted just a few decades after the diptych, when the 'proper' renaissance was just getting underway. A small panel of the Flagellation in the Frick Collection, New York, is a famous 'problem' painting, whose attribution has been disputed between Siena (Duccio or his circle) and Florence (Giotto/Cimabue and their circle). It is an important debate for clarifying our understanding of this period, crucial in the history of art. Miraculously, in 2000 a related panel was discovered, undoubtedly from the same altar and undoubtedly by Cimabue given its close relationship to documented works. Both panels are complete scenes in themselves, but they are parts of a small multi-panel altarpiece (polyptych) and relate closely to each other. The London panel also clarifies our understanding of its New York counterpart, which has been so controversial.

Finally, a Bernardo Daddi Coronation of the Virgin is shown with a small panel, 'Four Musical Angels', which was sawn off and is now at Christ Church, Oxford. Daddi was a student of Giotto, and was relatively prolific with a large workshop assisting him. This large panel has the powerful presence of earlier works by Giotto and Cimabue, but their austerity is tempered by a decorative sensibility that points forwards to the International Gothic style represented by, for example, Gentile da Fabriano. The two panels don't relate as easily as the others here, because the whole panel has been trimmed down with losses, and are now differently framed. But the accompanying booklet re-creates the context effectively, and seeing the two parts together makes it easier to appreciate the whole.

Each of these acquisitions fills a gap in the National Gallery's collection of early Italian art - remarkable given both the paucity of such works in private hands, and the extremely comprehensive collection at the National Gallery. Few very early Italian paintings of any importance have survived, and the Umbrian diptych is an important example from this obscure period.

There were no certainly attributed works be either of the two most celebrated 'founders' of the Renaissance in Florence - Cimabue and Giotto - until the acquisition of the Cimabue. It was regarded as so important that the director of the National Gallery is said to have warned off other likely purchasers - the Getty Museum in Los Angeles and the Metropolitan Museum of Art in New York - with the threat that an export license would be denied. And Bernardo Daddi is one of the few major Florentine Renaissance artists unrepresented in the National Gallery's collection, although there are two (a triptych and a polyptych) just down the road at the Courtauld Institute. In the National Gallery it provides a bridge between the earliest Renaissance artists, concentrated in Florence and Siena, and the constellation of artists that succeeded them.

This highlights a dilemma. These panels relate to other parts of the same altarpieces, but they also relate to other paintings that are shown in galleries as part of a coherent collection showing the development and dissemination of Italian Renaissance art as a crucial opening chapter in the story of Western art. There are unavoidable choices between the integrity of the individual work of art, and the integrity of collections as a whole. There just aren't enough early renaissance paintings to go around.

In 1945 Thomas Bodkin, Director of Birmingham's Barber Institute, argued for an international commission to broker the 'reunion' of fragmentary paintings and dismembered altarpieces. He cited the precedent of the Treaty of Versailles, which provided for Van Eyck's Ghent Polyptych being reunited after World War I. Since then, we have gone backwards - returning individual paintings to people with legal claims over them, or tribal objects to groups who claim spiritual affinity with them, but neglecting aesthetic criteria that might justify a more coherent approach to ownership claims over artistic treasures. Today few would justify further sub-division of composite works (the separation of Warhol's multiples, perhaps?), but equally few argue for permanent reunion.

Of course, the obvious solution is to do nothing, which will avoid difficult arguments altogether. This is cowardly and intellectually untenable. Granted, there are practical reasons for maintaining the status quo, given the difficulty of persuading galleries to give up their treasures and the legal rights of the owners. But tribal objects and looted art has been returned, so it is at least possible that if the argument is won, museums could sign up to some arrangement for permanent 'reunion'. In any case, can't does not imply shouldn't.

A more compelling argument is that the story is more important than the art. Comprehensive collections are more than the sum of their parts, because they show the interaction of artistic movements, the development of different styles, and contrasts between different schools and periods.

This is true, but one-sided. No gallery can be truly representative, and quirky museums with particular strengths and weaknesses have special appeal. Some periods of immense importance for the history of art have produced relatively few paintings (or at least few survivors), such as early Netherlandish art. Some kinds of art have had especially poor survival rates, such as renaissance paintings on linen, or processional shields (only one is extant). Others are highly concentrated, such as German renaissance art in German museums, or Spanish art in Spain.

Frescos are almost entirely absent from museum collections, and fortunately the argument has not been taken to the conclusion that we should rip them from the walls. But this is the logical extreme of this argument, because it means refusing to accept the arbitrary distribution of art. This reductio ad absurdum can be resisted, though, if we temper it with the respect for individual works of art. In a sense, this has to be given in the defence of the universal museum - the parts must have some inherent worth. But this is sometimes forgotten, and the individual works belittled by the overbearing context. This is particularly evident in some museum displays, where bizarre and jarring displays make it harder to appreciate individual works, and storyboards take up more space than exhibits.

I second Bodkin's argument - wherever possible, coherent works of art should be brought back together, so that we can better appreciate them as a whole. Applying this to the most celebrated case - the Parthenon Marbles - I would argue for bringing the parts back together. However, given that the larger part is in the British Museum, and given that it is not desirable to put it back on the Parthenon, exposed to the elements, they should be brought together in the British Museum. A museum holding several parts of a single altarpiece would never show only some panels - a coherent work of art is an unbreakable context. My claim is that the coherence of a work of art outweighs the importance of a comprehensive museum collection. The claims can be balanced by careful brokering of claims over reunited works, and the losses that some museums would suffer would be fully justified by the gain of seeing works of art back together, as they always should have been.

Bodkin's compelling argument has never been refuted; it has just been ignored. No one is speaking up and calling for the 'reunion' of paintings that never should have been separated. But this exhibition shows the urgent relevance of his case. These paintings should be kept together.

Till 29 January 2006