Stringed instruments have probably been around since the first time someone stretched a gut, rawhide, or fiber string over a resonator and plucked it. These early prototypes evolved over time into differenti­ated, often elaborately decorated and revered instruments. The University of Pennsylvania Museum is fortunate to house some of the earliest actual remains of stringed instruments from the Near East: intricate and beautiful examples from the Royal Cemetery at Ur in southern Iraq, dating to about the middle of the 3rd millennium BC (see Kilmer, this issue). Among these is the unique “boat-shaped” lyre, with a silver stag adorning the front.

Several of the grave pits in this cemetery and a few of the tomb chambers—almost all of the latter had  been looted in antiquity—had musical instruments among their grave goods (Fig. 1). These were perhaps intended to provide melodic accompaniment for the dead. The general shapes of the instruments were pre­served through their precious-metal sheathing or deco­rative inlays of non-perishable materials, or in some cases, their outlines were identified by the voids left by the perished materials from which they had been made. Nine lyres and two harps were found (see box on Lyres and Harps Compared), as well as other instruments including a pair of pan pipes and perhaps some sistra (rattles).

The Boat-Shaped Lyre as Found

The boat-shaped lyre was excavated during the 1928-29 season in the so-called Great Death Pit (desig­nated PG 1237), the same grave that yielded the “Ram in the Thicket” (see Rakic, this issue, and Rakic Fig. 3). It was found with two box lyres now in the British Museum, London, and the Iraq Museum, Baghdad, and near a pair of badly corroded figures of copper roe deer standing in trees and set on a rectangular base (see Fig. 1). The boat-shaped lyre lay on the face that is now its “front.” The figure of a silver stag (identified as a roe deer, native to the northern grasslands rather than the southern alluvial plain of Ur) stood with his front hooves supported by the branches of a copper tree, just like the goat in the “Ram in the Thicket.” His rear hooves stood on the top of the soundbox and conformed to its width, while the front upright of the instrument passed between his horns. Like the “Ram,” the stag’s head was flanked by two branches, with a central branch at his chin. Unlike the “Ram,” however, each branch was tipped by a single spade-shaped leaf.

The wooden soundbox had been covered with a single silver sheet which ran unbroken from side to side over the base, while a separate narrow strip completely covered its top. This sheathing and other non-perish­able parts were all that survived and defined the lyre. Although Woolley carefully and accurately recorded and described the lyre as he excavated it, its unique shape and the equally unique presence of the stag soon raised questions. The rounded soundbox and randed shape of the rear upright resembled those of a harp; while the similarity of the stag to the two copper stags, as well as its use in the context of a musical instrument, raised the question of whether the stag was part of the instrument. Before the new conservation started in 1994 the issues of whether this was a single instrument or parts of two and the inclusion of the stag had already been addressed, but the new study, using modern analytical techniques, provided an opportunity to confirm the previous con­clusions.

When found the lyre was in such brittle and fragile condition that it had to be conserved simply to lift it from the ground. According to Woolley, “The [sil­ver] metal was not only cracked into innumerable pieces but was so completely reduced to chloride that it was in many places no more than powder which had to be solidified with wax” (1934, pt. 2:122). Bandage, plaster, and wooden supports were also used to stabilize the lyre. After additional conservation work both at the site and at the British Museum, including adding metal supports to the soundbox and stag and carefully placing the lyre in a plaster surround, it was shipped to the University Museum in 1930. There it remained on display until 1979, when deterioration of the restoration required its removal and disassembly (Fig. 3).

An Opportunity for Restudy

In 1994, with conservator Tamsen Fuller of Northwest Objects Conservation Laboratory, we embarked on the long process of conserving and study­ing the boat-shaped lyre to prepare it for its return to exhibition. As well as restabilizing the fragile remains and removing unfortunate cosmetic reconstructions imposed on the tree, the work offered an opportunity to restudy the lyre completely. The reverse of the lyre, the first side seen by Woolley, had not been visible since the instru­ment was placed in the plaster surround.

One of the goals of the restudy was to accumulate as much new information as possible about ancient lyres. Little is known about the tuning or sound of these instruments other than from descriptions in later texts (see Kilmer, this issue). We hoped to learn from the physical remains of the lyre something about the materials from which the instrument had been made, how it was strung, and how its sound was shaped, produced, and projected. To this end we took advantage of several technologies that were unknown at the time of excavation to try to add to the information in Woolley’s original careful description. Many scientists and scholars provided their specialized knowledge and services.

As a result of the new study, most of the conclu­sions which had been previously drawn, particularly those related to what parts belonged together and the stringing, were completely confirmed: all the parts came from one, not two, instruments; it had been strung from the yoke, not the rear upright, to the soundbox; the stag was part of the lyre (de Schauensee nod.).

How the Lyre was Made and Strung

The interior of the silver sheathing was careful­ly examined in an effort to gain information about the construction of the soundbox. The configuration of a stringed instrument’s resonator, as well as the wood from which it is made, will affect the type and quality of the sound produced. Unfortunately, only the fact that the soundbox had been made of sheets of wood with the grain laid horizontally could be determined from the remains; this was all that was preserved in residual markings found in the silver chloride corrosion products of the sheathing (Fig. 4). The structure of the few frag­ments of wood found during conservation of the instru­ment was not well enagh preserved, nor free enough from the wax used in the field conservation, for identifi­cation. Compositional analysis of the silver sheathing conducted by proton-induced X-ray emission (PIXE) spectroscopy showed it to have been 99 percent pure.

Examination of the sheathing confirmed that the soundbox had no opening other than the narrow slit on one side where the strings had been attached. This opening was not large enough to serve as a sound hole from which the vibrations created within the box could emerge (Lawergren and Gurney 1987). It has been suggested that the bottom of the soundboxes of box lyres, such as the Univer­sity Museum’s bull-headed lyre and the British Museum’s silver lyre, were left open so that the resonant sound described in later texts could emerge. While this is possible, the folded edge of the sheathing wrapped over the narrow base of the boat-shaped lyre’s soundbox shows that such was not the case here. Direct evidence of how the sound was pro­duced or what type of sound was favored therefore remains elusive.

There was no evidence of a bridge on our lyre, but one would have been required for play, and the impression of a bridge was found on the silver lyre in the British Museum. Markings of the strings were clear­ly preserved in the silver chloride above the slit, indicat­ing that they were attached here (see Kilmer, this issue, for a discussion of how this might have been done; also Kilmer’s Fig. 9). At the yoke, however, only the marks of the wrappings that would have been placed under the strings were preserved (Fig. 5a). Macroscopic pho­tographs suggested that the wrappings over which the strings were wound may have consisted of a sort of loosely braided or criss-cross wrapped plant fiber, such as is still sometimes used for lyres in some parts of Africa, rather than woven cloth (Fig. 5b). Unfortunately, attempts to study the markings on the yoke at much higher magnifications produced ambiguous results.

Questions Answered, Questions Raised

X rays and CAT scans were obtained to provide information about previous conservation efforts, the condition of the preserved silver, and the construction of the lyre. What was clearly revealed by conventional radiology was how the silver sheet was used and attached. Each part of the instrument—the uprights, the yoke, and the body and top of the soundbox—had been separately covered with silver sheet which, with the exception of the yoke, was held in place by identical small tacks of silver or silver alloy (Fig. 6). No tacks were found in the yoke. Here, the edge of the sheet had been folded over onto itself and hammered into place. A separate cap had been fitted over the end of the yoke and secured by bitumen.

Some X rays raised questions, however, such as the revelation of apparent staple-like tacks in the rump of thee stag, but these were subsequently clarified by CAT scan images (Fig. 7). CAT scans also provided the completely new information that the stag’s head had been modeled at of bitumen over an armature of cop­per rods and pins (Fig. 8a, b). This method of using bitumen is of particular importance because althagh its use as a filling or as a support for hammered sheet metal was well known, both at Ur and elsewhere, its use with an armature was previously unknown.

The body of the stag had been made of wood in individual sections that were later joined to each other.

While a single heavy copper nail runs through the rump of the animal, its purpose and the method used to attach the legs to the body remain unknown. Mortise and tenon construction, as suggested by Woolley for assem­bling the parts of box lyres (1934, pt. 1:257), may have been used here as well. Each section was covered with silver sheet held in place by tacks before assembly (Fig. 9a, b). This is a technique also seen in other pieces from the Royal Cemetery now in the University Museum, including both the gold sheet–covered head and the sil­ver-backed beard of the bull-headed lyre and the gold-covered sections of the goat in the “Ram in the Thicket” (see, respectively, Kilmer and Rakic in this issue).

While the technical studies were being carried out, conservation work proceeded: removing the old wax, bandage, plaster of Paris, and other materials, cleaning the surfaces, and replacing fills and supports with inert materials. Fragments of wood found during conservation of the stag figure were identified as proba­bly belonging to the box (Buxus) or pistachio (Pirtacia) family, neither of which was native to the alluvial plain of southern Mesopotamia, but rather to areas further north. Meticulous records of the conservation and the technical findings were made, in written records, digi­tized images and schematic renderings, slides, and black and white photographs. Finally the lyre was returned to exhibition (Fig. 10).

Most of what we know about music, musicolo­gy, and music theory comes from later texts, while information about the instruments themselves comes largely from representations and texts. That from actual early instruments is rare, and our newly gained knowledge supplements the meager amount of information previ­ously known. As we expand our knowledge, our respect for the skill and expertise of the early craftsmen who made the instruments, played them, and wrote the music for them increases as well (Fig. 11).


The author would like to thank Dr. Robert Dyson for making the undertaking of this project possible, Dr. Stuart Fleming for his assistance with the many technical studies, Dr. Charles Swann for the PIKE studies, Dr. Naomi F. Miller for wood identification, Dr. Anne Kilmer for her helpful thoughts, and the anonymous readers of this article for their useful suggestions. The author also thanks the Hospital of the University of Pennsylvania for providing X rays and CAT scans, in particular Ann Rufo and Dr. Bernard Birnbaum of that institution for their generous assistance with the radiology and its interpretation.