INSTITUTE OF NAUTICAL ARCHAEOLOGY - Bodrum, Turkey |
| ULUBURUN |
The Copper Oxhide and Bun Ingots
by Dr. Patricia Sibella
Of the 354 complete copper oxhide ingots and 121 bun ingots found on the Uluburun wreck, 81 were cleaned, drawn, recorded, and examined between June and November of 1995, bringing the total that have been studied to date to 180. Particular attention was devoted to the modification of the oxhide ingot typology first established by H. G. Buchholz, and later modified by G.F. Bass in his study of the 34 examples from the Cape Gelidonya shipwreck. There are 31 ingots of a type unique to Uluburun, provided with only two handle-like protrusions, one on each corner of one of the ingot's long sides. We have designated these as type 4. For the purpose of our research, the traditional type 3, as well as type 4, have each been divided into 4 subtypes. These subdivisions are primarily based on the ratios of the ingots' dimensions and on the shape of the handle-like protrusions.
fig. 1.Marks founded in the Oxhide ingots from Uluburun
In addition to the morphological features of these ingots, we have documented and studied the marks chiseled into their surfaces. Although most of the ingots still need to be cleaned of surface encrustation, preliminary examination has revealed that at least half of the oxhide ingots, or about 160 examples, are incised with at least one, and possibly as many as three, marks on their upper or rougher surfaces opposite their mold sides. In addition, six oxhide ingots bear linear incisions along their shorter edges, but only three of these six examples are also associated with marks on the upper surface. In contrast, only 28 out of the 121 bun ingots appear to be marked. This is still a high percentage, considering that (so far as we know) no marked bun ingots have been found in any other land or underwater site. On the Uluburun bun ingots, these marks are always single marks on their lower, or smoother mold sides. An exception is KW 1088, which shows an incised mark on its rough surface.
In some cases, the incisions are preserved sufficiently to document the individual chisel strokes in the mark, thereby allowing us to determine not only the sequence in which the strokes were made, but also the shape and size of the chisel used and the angle at which it was struck. Close attention should be given here to the traces of wear discernible on the chisels found on board the ship to determine if some of these tools may have been used to mark the ingots. If so, this discovery may provide some information about when these marks were made.
The 64 marks thus far examined on the surfaces of oxhide and bun ingots comprise only 32 different shapes. Of these marks, 13 appear more than once (fig.1: 1b,1d, 2c-d, 3b, 4b-d, 5a, 6c-d, 7c-d) and one is repeated at least six times (fig.1: 4d). Some marks are common to both the oxhide and bun ingots, but there are fewer types of markings on the latter variety of ingot. Of the six types of marks observed on the bun ingots, five are also found on the oxhide shape (fig.1: 2d, 3b, 4b, 4d, 6d). The sixth mark, however, is found only on the oval bun ingots and appears on all six of them. The precise location of the marks on the surfaces of the oxhide ingots, and the diverse mark combinations, do not appear to follow any specific pattern. The specific marks cannot be associated with certain ingot subtypes. The marks vary in complexity from a simple cross (fig.1: 5a) or a fishhook (fig,1: 1c) to a complicated trident (fig.1: 2b), a fish-like shape (fig.1: 7b) or a sailing boat (fig.1: 5b). One of the Uluburun marks (fig. l: 3a), also seen on a copper oxhide ingot from Ayia Triada in Crete, seems to find its parallel in the later 11th-century B.C. northwestern Semitic syllabary. In only one instance does a perfectly identical mark, almost certainly made by the same hand (i.e. same orientation of stroke, depth of incision, size of tool, etc.) appear on two separate oxhide ingots (fig.1: 2c). Each of the two ingots thus marked also has a second V-shaped mark chiseled along one of its shorter edges. These two ingots do not belong to the same ingot subtype, but future studies may reveal identically-marked examples on the same subtype. Any markings of this nature may have profound implications for our understanding of ancient metallurgical practices and the mechanisms by which these ingots were distributed.
[top]
Although similar marks are also observed on oxhide ingots from Ayia Triada in Crete (fig. 1: 3a, 6a), San d'Antioco di Bisarcio in Sardinia (fig. 1: 6d), and Enkomi in Cyprus (fig. l: 6d, 3c), their meaning and purpose still elude us. That the Uluburun marks were incised and not stamped (as were the many examples from Cape Gelidonya) leads us to believe that these marks probably were made at some point of receipt or export rather than at primary production centers. This supposition seems to find further confirmation in a similar mark that appears on a tin ingot (fig.1: 7a). As tin and copper are mined in different geographical regions, it is highly unlikely that the same mark was placed on ingots of dissimilar metals and of diverse origins, unless they were incised at a center that initially received both metals before they were shipped.
Concerted efforts have also been dedicated to the identification of mold siblings, ingots cast in the same mold. This is an acknowledged but poorly attested phenomenon in the archaeological record that could help us better understand Bronze Age casting techniques. The mold impressions of at least five ovoid bun ingots, bearing identical incised marks (fig. l: 7d), demonstrate that they indeed are mold siblings, as are two pairs of small type of oxhide ingots studied and presented earlier. It seems likely, therefore, that these ingots were cast in reusable stone or clay molds, rather than in perishable sand molds. We have yet to discover siblings of full-sized oxhide ingots. There are indications that these ingots may have been cast in multiple pourings of molten metal into the mold. An indication of this practice may be observed on the edges of oxhide ingots that frequently display what we have provisionally termed as a "casting groove." Presumably, this groove results from the lower half of the ingot being poured first and contracting slightly upon cooling before the second pour was made. This groove, on the other hand, usually does not appear along the perimeter of bun ingots, which suggests that they were cast in single pourings.
Until the remaining Uluburun ingots are cleaned, our comprehension of these marks will 'remain inconclusive. Of interest in this connection is our continuing study of the distribution of the incised marks and their association with ingot types and subtypes. We are also searching for possible patterns in the ingots' stacking within the ship's hold. Such ingot clusterings, if they occur, may provide clues about whether the copper ingots were stacked in specific batches that have common origins, ownership, and quality of metal. Lead-isotope analyses are being conducted by Noel Gale and Sophie Stos-Gale of Oxford University Isotrace Laboratory on 80 copper ingots recovered from the Uluburun shipwreck, and 45 ingots from the Cape Gelidonya shipwreck. The pending results of these analyses may help us to answer some of the foregoing questions.
Lead-isotope analysis is also imperative for determining the source(s) of copper from which the ingots were most likely cast, and establishing possible correlations between copper source, ingot types, and incised marks. Samples for lead-isotope studies were also taken from 71 other bronze and copper objects from Cape Gelidonya, and 14 tin ingots from Uluburun. This may enable further correlations between the sources of the ingots and the sources of the metal in the artifacts.
[top]
