Historical Text and Historical Object: The Poetics of the Musee de Cluny

Abstract

The story is told by the British art historian Lord Clark that, in the days when he used to stay at Berenson's Florentine villa, I Tatti, he would try the experiment of moving a small Renaissance bronze a few inches from its original position each evening on retiring to bed. Each morning, as he came down to breakfast, he was able to note that it had been restored with great precision to its former location. This story illustrates, of course, more than a mere mania for domestic order. To Berenson, no doubt-if not to Lord Clark the bronze was not simply an object which could be moved here and there without detriment to its intrinsic aesthetic significance. It was a term in a system, whose exact relationship to other terms had to be maintained as, by imperceptible stages, Berenson's home became the Berenson Museum. Yet when, with Berenson's death, that process had become complete, we may well wonder how much of this original order was in fact preserved for posterity. Given that the original placing of each object within a defined series of contiguities was indeed the result of his intentions, we might ask if these intentions were likely to be communicated to the new stewards of his collection in the form of a precise system. If not, discounting the possibility of an intuitive rapport beyond the grave, we would be bound to envisage the inevitability of the museum collapsing into a form of anomie. Borrowing the terminology of linguistics, we would say that the systematic plane being barred by Berenson's absence, the syntagmatic plane (the ordering of objects in real space) would inevitably lose its coherence. If the museum continued to communicate, it would at the same time be afflicted with a chronic aphasia. Connections and relationships which were once the visible demonstration of a total view of art and the world would have been reduced to mere contiguities and juxtapositions.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 9 citations exist upstream.