The lost-wax technique offered ancient bronze-workers opportunities for both replication and variation. Evidence for the process has frequently been documented, but the careful and nuanced planning that occurred when working with a wax model deserves closer scrutiny. Through a close collaboration with the Museo Archeologico Nazionale di Napoli, we have examined three bronze sculptures from the Bay of Naples - the Apollo and Diana from Pompeii (MANN Inv. 5629 and 4895), and the Tiberius from Herculaneum (MANN Inv. 5615). All offer evidence for adjustments in the wax that benefited the casting of the bronze and its assembly.
Our first project involved two statues from the same workshop (Eric Risser and D. Saunders, "The Bronze Apollo and Diana from Pompeii: an Example of Serial Production." AK 58, 2015, 78-96). The limbs of the Apollo and Diana derived from shared models, but showed signs of shaping during the wax-working stage, suggesting that they had required adaptation for each figure. Stepped joins or a notch-and-tab system facilitated the orientation of parts after casting. Particularly unusual was the void in the back of the Apollo. Comparable, albeit smaller, voids were noted on the arms. These had all been created in the wax and were subsequently hidden by sections of drapery. They appear to have been made to facilitate the casting and piecing processes, and indicate careful planning across the different phases of production.
Our study of the Tiberius allowed us to develop this theme further (E. Risser and D. Saunders, "Tiberius from Herculaneum: Methods of Assembling a Monumental Bronze Portrait," in Proceedings of the 19th International Congress on Ancients Bronzes, Los Angeles: Getty Publications, forthcoming). We fully expected the monumental figure to have been assembled from numerous different parts (as demonstrated for the other large bronzes at Herculaneum by Götz Lahusen and Edilberto Formigli (Grossbronzen aus Herculaneum und Pompeji, Worms: Wernersche Verlagsgesellschaft, 2007), but were taken aback to count 62. This seems too many to be practical, requiring intensive labor to divide up the wax model for casting. We propose, however, that such effort was worthwhile, for it facilitated the later - and most challenging - phases of production. The figure's complex drapery forms were simplified into flat or openly curved sections that could then be cast with minimal risk. Once these parts were cast, the portrait was pieced together with just a small number of joins. As we had observed when studying the Apollo and Diana, quick tack welds were often employed in assembling the figure. Where seams were difficult to see, they were left largely open.
The opportunity to study these three statues in detail provides a new appreciation for the joined-up thinking entailed in the production of bronze sculptures efficiently and with minimal risk. The recent Power and Pathos exhibition at the J. Paul Getty Museum offered a chance to look closely at several other ancient bronzes. Our preliminary observations suggest that the forward planning detailed for the Apollo, Diana and Tiberius was a well-established way of working in ancient bronze workshops.
Herculaneum: New Technologies and New Discoveries in Art and Text