Main Menu · Search · Current Issue · Contact · Archives · Centennial · Letters to the Editor · FAQs

But the vase cannot tell the whole story. ancient bronze founders used a casting method known as the lost-wax process, in which a wax model was replaced by molten bronze. After heating the bronze to a liquid state, just over 1,000 degrees centigrade, the remaining challenge was to substitute the molten bronze for the wax. This could be done by two separate methods; the differences between them are often blurred because ancient technicians combined elements of both.

In the direct lost-wax process, an object was first roughed out in clay. Then the artist had to fashion a perfect wax model over the clay, which is known as the core. Iron pins called chaplets were sunk through the wax and into the clay core material. With the ends of the iron chaplets still projecting outside the wax model, a clay mold was applied in layers to the wax. Once the outermost layer of clay had become thick and dry, the entire mass could be heated, during which process the now liquid wax was poured or burned out. The iron chaplets remained, holding the inner clay core and the outer clay mold in static relation to one another. Then bronze could be poured into the space left by the wax and, once the metal had cooled, the outer mold broken away, exposing the bronze. Since both the clay mold and the wax model were destroyed, the resulting bronze was unique. Says Mattusch, "Almost nothing of life size was made this way. It was too risky."

In the far more common indirect method, a mold was taken from a model. The model was often made of clay, but it could also have been a well-known bronze or marble statue that someone wanted to copy. Once the mold had been made, it was lined with a thin layer of wax, making an impression, or working model, that exactly replicated the original. After the wax model was removed from the clay mold, artisans carved into or added to the wax fine details like hair, beard, and fingernails. When the wax model was finished, metal workers filled it with a clay core, pushed chaplets through the wax into the core, and covered the assembly with an outer clay mold for baking and casting. When the casting was completed, the original model and the first clay mold taken from it remained intact. The process could be repeated ad infinitum.

All but two of the bronzes in The Fire of Hephaistos are products of the indirect lost-wax method. This is because ancient bronzeworkers preferred the indirect process for casting large bronze statuary. Indirect casting allowed them to create multiple copies of a statue. If a mistake was made during the indirect casting of the arm or leg of a figure, a new limb could always be made from the original mold. Futhermore, a single model could be used to create several slightly different versions of, for example, a single Aphrodite. The difference from one casting of the goddess to the next was in the details added and adjustments made to the wax working model.

Clearly, classical bronze manufacture was not the ancient forebear of Henry Ford's model T production line. "A limited-edition statue by Auguste Rodin would be a better analogy," says Mattusch. Accepting even that comparison is a leap for many people. "The way two bronzes that were just alike have been studied before, was that art historians would say `these must be two different gods, or Castor and Pollux, or the legs of a table: two things that belonged together.' Experts on the technology would look at them and say `Oh, obviously it's a series casting, probably from a production line in a workshop.' Then we find the bronze alloys are exactly the same," says Mattusch, "and that clinches it.

"What we're looking at is what people don't like to admit about classical bronzes: that we're looking at two versions of the same thing," she continues. "The reason we didn't start with this understanding in the first place, was that we found these antiquities one at a time. We didn't find a workshop that had 10 of the same thing in it. We would find one Aphrodite and say, `God, it's a masterpiece.' And then we'd find another one someplace else, and instead of asking, `Is this the same workshop' or `Is this two editions of the same thing,' we'd say, `Same artist' or `Same school.'"

On the other hand, Mattusch points out that just because two statues look similar doesn't mean they are from the same workshop. Three statues of a goddess in the exhibit are a case in point. Despite the fact that they "are just alike," according to Mattusch, "they are probably three different editions coming from three different places, because this goddess was a popular type." How can she be so sure? Because the three have different alloys. "Metal was tested from all the statues for the first time for this show," says Mattusch, "and all the samples were tested at the same time and in the same way." That did settle the question of provenance, but not before the samples had made an unplanned trip to Hawaii.

Henry Lie uses a portable drill to take small samples from unobtrusive spots in bronze statuary. The bronze drill-ings are then turned over to Suzanne Young, of Harvard's archaeometry laboratories. Young is a chemist who does research on bronze, malachite, turquoise, gold, and copper, and her analyses will identify the precise elemental makeup of each bronze sample. It is perhaps a sign of the burgeoning interdisciplinary cooperation at the University that, although Young has been doing this kind of analysis for 13 years, this is her first experience working with an art museum. She weighs out 25 milligrams of each drilling, dissolves them in quartz-distilled nitric and hydrochloric acid in volumetric flasks, calculates the dilution factor, and then lets the mixture stand for 24 hours. The result is a viridescent potion that ends up looking like lime Kool-Aid. Young measures an aliquot from each flask, just enough to run through the sensitive inductively coupled plasma quadrupole mass spectrometer (ICP-MS), the heart of which is a brilliant purple argon gas plasma torch (8,000 degrees centigrade) that ionizes and atomizes everything it touches. Visitors to the lab are warned not to look directly at the light. The spectrometry is scheduled so that Lie and Mattusch will have plenty of time to study the data. But Harvard's ICP-MS won't run properly, so Young's aliquots must be sent to the department of geology and geophysics at the University of Hawaii. This sudden winter vacation doesn't harm the samples, but it does mean delayed results; once the elemental analysis has come back, Mattusch and Lie do a last-minute rewrite of many of the catalog entries for bronzes in the exhibit.

The ICP-MS data gives the exact percentage of every trace element in each bronze, and Mattusch suggests that near-perfect correspondences in the trace elements of two pieces allegedly from the same region in northern Syria support their provenance. Lie relies on a complementary analytical technique called electron probe microanalysis (EPMA), conducted at Harvard's department of earth and planetary sciences, to learn the precise bronze alloy "recipe" of copper and tin for each statue. Together, ICP-MS and EPMA provide a complete elemental profile of each sample. The differences in the ratio of copper to tin among the three similar goddesses in the exhibit allow Mattusch to ascribe separate origins to these statues.

EPMA analysis also reveals that five of the bronzes contain areas of pure copper, and additional areas where tin and lead are poorly integrated. These are the main ingredients of bronze. The fact that they never mixed properly suggests to Young and Lie that "these bronzes were cast at very low temperatures." Were the ancient artisans only just able to heat their furnaces to high enough temperatures? Or were they rushing the process to finish these bronzes for the next customer in a line of eager buyers? Says Mattusch, "We are just now learning what sort of questions to ask."

Main Menu · Search · Current Issue · Contact · Archives · Centennial · Letters to the Editor · FAQs