Secret processes, lost technologies and materials that survive by name alone have an irresistible attraction – especially if they are metals. Records from the past describe many processes and materials whose identity we can only guess at. Just occasionally, however, it is possible to match ancient descriptions with artefacts that have survived from antiquity and now lie in museums, their true origins unrecognised. It now turns out that some of the most prestigious metals from the ancient West are almost identical to a sophisticated metal, known as shakudo, that is still made in Japan. The trail led from Egyptian hieroglyphics of 3500 years ago to clues contained in Homer’s Iliad, on to the famous and highly prized Corinthian bronze of the Roman Empire, eastwards with more clues from the great epics of Indian and Tibetan literature, then to the work of barely remembered Chinese craftsmen before culminating in 19th century Japan.
The detective work took place at the British Museum in London, where we had suspected that some of the processes briefly described by ancient alchemists for making a special type of bronze with a black patina were similar to those for making shakudo. By a happy combination of detailed examination of metalwork, careful study of ancient literature, some intuition and a little luck we were able to turn promising hypotheses into fact, and build up a cultural and technical trail stretching over half the globe through more than 3000 years.
Shakudo was probably first used in Japan in the 14th century, but seems not to have been known in the West until the mid-19th century when imperial rule was restored and Japan opened its doors to foreigners for the first time in more than two centuries. Among the most impressive artistic creations early travellers brought back were sword fittings, or tsuba, made of shakudo. These are copper with small amounts of gold, and sometimes silver and arsenic, which are inlaid with gold, silver and other metals, and then chemically treated. The bronze alloy develops a deep rich purple-black patina, providing a perfect background to the shiny inlays.
Advertisement
Shakudo, with its black patina and inlays of gold and silver, quickly became a collectors’ item in an enthusiasm for all things Japanese. However, exact details of the composition and manufacture of shakudo were not known in the West until observed by a number of Europeans living in Japan. The most eminent of these was William Gowland, a metallurgist who had gone to Japan in the 1870s to help set up a modern mint. Gowland took a keen interest in all aspects of traditional Japanese metallurgy, especially the shakudo alloys. When he returned to England in 1888, he lectured extensively and wrote about these exotic alloys.
The processes observed by Gowland and practised to this day in Japan have many variants, but the essentials are the same. After the alloy is made and cast to shape, it is inlaid with gold, silver or other metals, and carefully polished with a variety of ever softer abrasives, including charcoal made from burned deer antler. The surface is then cleaned and degreased using natural organic reagents such as the juice of the Japanese radish or of the bitter plum.
After this preparation, the inlaid shakudo is ready for patination. This is done by treating the metal in a hot, mildly acidic aqueous solution containing small amounts of a variety of ingredients such as verdigris (copper acetate) and blue vitriol (copper sulphate). A superb velvet black or deep purple patina soon develops, while the metals of the inlays remain unaffected.
The exact chemical composition of the patinated layer is unclear. The most common black copper mineral is tenorite (CuO) and one would expect that to be present; X-ray diffraction, however, reveals only cuprite (Cu2O), a red oxide. Michael Notis of Lehigh University, Pennsylvania, suggests that silver oxides and gold particles play a part in the formation of the colour and texture. Silver oxides can have a purple hue and the minute particles of gold may interrupt the regular crystalline array of the cuprite, changing its appearance.
For many years, and with little or no evidence to go on, historians presumed that the Japanese learnt of the alloy from the Chinese in the 14th century. No one had any reason to suppose that the shakudo-type alloys had a long history in the West. But the reference to the juice of wild plum used prior to patination gave us an early clue that similar processes might have been used in the West thousands of years before. The plum is a source of acetic, formic and malic acids. The Roman alchemist Zosimos, who lived in Alexandria in the 3rd century AD, described a process, known as iosis, in which a metal was treated with rhubarb so that it would develop a purple patina. Rhubarb is also a source of acetic, formic and malic acids.
Hieroglyphic clues
This led us, together with Alessandra Giumlia Mair, to examine other ancient literature of the West where there are frequent references to inlaid black bronzes, such as the hsmn-km (literally black copper) of the Egyptians, or to black gold such as that described in the Iliad. In this epic poem, Homer tells of the shield that Hephaestos, the Greek god of fire and the forge, made for Achilles by throwing copper, tin, silver and gold into the furnace. Homer also describes the decoration on the shield, and states that a ploughed field was portrayed in black, although it contained gold. Similarly, there are many references to the highly prized Roman alloy, Corinthian bronze, that was said to contain small quantities of gold and silver. It seemed possible to us that at least some of these could refer to shakudo-type alloys.
About 30 years ago, J. D. Cooney, an Egyp-tologist at the Brooklyn Museum in New York, had drawn attention to some bronzes, found in museum collections around the world, that had a deep black patina and were inlaid with gold. He suggested that these were the hsmn-km bronzes referred to on Egyptian hieroglyphs. These references start between about 1500 and 1300 BC, in the 18th dynasty, and always describe the hsmn-km bronzes as being inlaid with precious metals, usually gold. Cooney went on to suggest that the famous inlaid Mycenaean daggers were of a similar alloy. In these daggers, which date from about the 15th century BC, panels of black bronze inlaid with gold and silver were set into the sides.
Cooney unfortunately had no access to scientific facilities, or felt he could dispense with them, and so made his one big mistake. He confidently stated that the black surfaces were copper sulphides, brought about by treating the bronze with overripe eggs. Bad eggs and rotten science.
Since then Jack Ogden, an independent researcher at Cambridge, has examined a My-cenaean dagger of the same type and period. He found the surface of the inlaid black panel contains copper oxides, not sulphides. Analysis also revealed that the black bronze of the panel contains 1.7 per cent gold, 0.55 per cent silver and 0.5 per cent arsenic – vastly in excess of the amounts of these elements normally found in ancient bronze, but similar to the levels in the shakudo and hsmn-km alloys.
Egyptian shakudo
We came to a similar conclusion when we analysed a selection of black patinated and inlaid Egyptian bronzes, mainly dated to between the 8th and 2nd centuries BC, from the collections of the British Museum. X-ray diffraction revealed that the black surfaces on some of them contain cuprite, while X-ray fluorescence and atomic absorption spectrometry showed the alloys to be of bronze with small amounts of gold, silver and arsenic. They were identical in all respects to shakudo, with no trace of copper sulphides.
This was not the first time that such a mistake had been made. In 1887, a superb Egyptian statuette of Osiris was displayed at the Society of Antiquaries of London, resplendent with gold inlays and black patination. A Professor Middleton noted that: ‘The method employed to colour the bronze was probably much the same as that used by the modern Japanese bronze workers.’ So far so good, but then he went on describe how the Japanese applied sulphur to the bronze to produce a black copper sulphide. This explanation was completely wrong, but Gowland was still in Japan. Had he been in the audience on that day he could have corrected Middleton and instigated some analyses. And the true similarity between Egyptian and Japanese metalworking techniques would undoubtedly have been established a century earlier.
Moving our investigations forward, we turned to the Romans who had a celebrated alloy known as Corinthian bronze – so celebrated that despite dozens of contemporary references, few authors bothered to say what it was. Few that is except the redoubtable Pliny, who gives a rather confused, and certainly mistranslated, account of it in the great Natural History published in 77 AD. After stating that it was the most prestigious bronze, and that no one knew how to make it properly any more (he tended to be rather pessimistic about his own times), he went on to describe different varieties of Corinthian bronze that were both alloyed and inlaid with precious metals.
Other clues occur elsewhere. The Roman philosopher Seneca, writing a few years earlier, derides those who collected articles made of Corinthian bronze, describing them as ‘rusty’, from which we can infer that the Corinthian bronzes were indeed patinated. The Greek philosopher Plutarch, who also lived in the 1st century AD, describes Corinthian bronze in a section of his Moralia devoted to corroded bronze, and notes that it was called bronze because there was only a little gold and silver present. Finally Pausanias, in his travelogue of Greece compiled in the 2nd century AD, states that in Corinth the bronzes were ‘dyed’ by being plunged red hot into the waters of the Peirene fountain which stood in the middle of the city.
The spring that fed this fountain still flows and by good fortune some American excavations in the old metalworking quarter of the city in the 1930s located great troughs fed directly from the spring – one of the few instances in which specific technical installations described in ancient literature have been located and confirmed by excavation. Analysis of the waters shows that they are unusually rich in minerals, which may have been of value in the patination process.
We have now carried out detailed scientific examinations of several Roman bronzes in the British Museum’s collection. Some are pieces of black bronze inlaid with gold or silver, while others are made of ordinary bronze inlaid with strips of black bronze. One such bronze is a statue of Nero as Alexander, dating from the 1st century AD and found in Norfolk last century. Archaeologists have always described the black material as niello – a black alloy of copper and silver sulphides used to fill designs cut into the surface of a metal – but the reality was very different. Analysis of the black inlays showed them to be of copper with small quantities of gold and silver, and with surfaces of copper oxides. It was Roman shakudo.
Analysis of the other black patinated and inlaid bronzes revealed that they usually contain a few per cent of gold and silver. This equates with the Corinthian bronze described by Pliny and the other Roman authors. The best evidence for this is a 15th-century Syriac translation of a text by Zosimos, now in the University of Cambridge Library. The document contains precise instructions for the production of Corinthian bronze: to be made from a mina of copper alloyed with 8 drachmas each of gold and silver (equivalent of about 6.5 per cent each of gold and silver) and then treated with a variety of solutions to develop the patina. The recipe is not dissimilar to the 19th-century Japanese accounts.
So, was this high prestige material introduced to the Orient from the West, or was it invented there independently? The old ideas of innovation spreading out from a very few advanced centres are out of favour with most archaeologists, but art historians recognise the influence of artistic styles between cultures as far apart as the classical world and China. We believe that Corinthian bronze could have been just such a cultural export.
Classic times
In support of this hypothesis are descriptions of purple bronzes in Indian and Tibetan classical literature. The Ramayana, the great epic poem of India, describes the priests officiating at a horse sacrifice as being rewarded with beans of gold coloured like damsons. This description of the horse sacrifice dates from at least the early centuries AD. The Tibetans seem to have had a similar alloy known as zi-khyim which, in a 10th-century account, is described as containing copper with iron, lead, tin, rock crystal, mercury and – gold and silver. The alloy was then treated with ‘poisonous water’ to produce a purple-red hue. In the 18th century, an English traveller called Samuel Turner described some of the bronzes he had seen in Tibetan monasteries as ‘resembling Mr Wedgwood’s black ware’.
Certainly Chinese records from the 5th century AD note an alloy tzu mo chin, literally purple-sheen gold, which Joseph Needham identifies as the missing progenitor of shakudo in his great Science and Civilisation in China series. Tzu mo chin has never been identified, but we have recently uncovered a very similar material.
And now the element of luck comes in. We were fortunate to have Mike Wayman of the University of Alberta working in our laboratory for a year. It transpired that a former colleague of his, H. B. Collier, had worked in the remote Chinese province of Yunnan as a biochemist in the 1920s and 1930s and collected some attractive black boxes, inlaid with silver. These were known as wu tong, literally black copper, which is apparently little known outside the province even though it was made there from at least the 15th century AD. Collier’s black boxes are the only known collection of wu tong outside China and, through Wayman, we were able to acquire some boxes. Analysis showed the wu tong to be an alloy of copper with a little gold, inlaid with silver and then patinated. Although the links between wu tong and shakudo are far from proven, we know shakudo alloys can form a perfectly acceptable patina with copper and a few per cent of gold.
We are slowly bridging the gap between ancient Egypt, Mycenae, Rome and Japan. It is possible that knowledge of this sophisticated and esoteric technology travelled East in the Roman period, either on the Silk Route across Asia, or possibly on a southern route via India and southeast Asia. The earliest Chinese references to dark or purple gold are found in some Daoist texts in about the 5th century AD, almost a thousand years before the first references to wu tong.
What of Corinthian bronze in the West? References continue right up to the end of the Roman Empire in the 7th century – and then silence. It turns out, however, that a thousand years before Gowland first described this quintessential Japanese alloy to London audiences, it was being made in the British Isles by Saxon and Celtic craftsmen. In the past few months, we have begun to identify the alloy among some quite homely Anglo-Saxon and Celtic Irish metalwork as late as the 8th and 9th centuries AD. These are in the form of black-headed studs set into enamelled bronze buckles and strap-ends. Only two or three examples have been identified so far, but we suspect that many more may come to light because black inlays and studs are not uncommon. Which leaves us with new questions: when was this sophisticated technique lost in the West, and by what name was it known in Anglo-Saxon England?
Paul Craddock is head of metals research at the British Museum, London. There is a small exhibition of some of these bronzes i Room 68 of the Greek and Roman gallery at the museum until March.
Further reading: Metal Plating and Patination, edited by S. La Niece and PT Craddock (Butterworth-Heinemann, 1993).
