15. Technological Study and Interpretation of Rhomboid Accessories from Grave Circle A, Mycenae1
p. 245-262
Remerciements
This paper is part of a research project on Mycenaean gold-working techniques, undertaken by the authors and partly funded by INSTAP. The experimental reconstruction of NAM 669A was generously supported, both financially and morally, by the Canadian Museum of History (Gatinaeu, Canada) and resulted in the production of a replica and a video for the exhibition The Greeks. From Agamemnon to Alexander the Great’ (Canada, USA, 2015-2016). We are grateful to both Institutes for their support and to the NAM for the permission to study thse objects. Wann thanks are also due to Ms Maria Kontaki and Ms Georgia Karamariou, conservators of the Metals Laboratory of the NAM, to Dr Yannis Bassiakos and Dr Liana Philippaki, of the Demokritos Research Centre, for the XRF analysis of NAM 346, to Dr Dimitra Mylona for the preliminary study of the bone cores of the Mycenae ‘buttons’, to the Photographic Archive of the NAM for the photos of Figs 15.1-15.3, and to Professor Joseph Maran for valuable comments on earlier drafts of this paper and bibliographic suggestions. Of course, any fallacies are the exclusive responsibility of the authors.
Texte intégral
1Among the luxurious artefacts that furnished the burials of Grave Circle A at Mycenae, Schliemann collected dozens of gold-coated items of bone, either of rounded or rhomboid shape, carved with geometric patterns (bands of meanders, concentric circles, spirals, whirls and zig-zags) (Schliemann 1880: 258-266; Karo 1930-33: pl. LIX-LXVI). Most of these objects date to LHI and have been interpreted as ‘buttons’, ‘brooches’ or generally ‘clothing ornaments’ (Vermeule 1975: 23; Dickinson 1977: 77, 85; Hiller 1991: 211). Rounded bone ornaments of the same type (but not gold-coated) are also known from the LH IIA tholos tomb A of Kakovatos, Messenia (Müller 1909: 283, 285-6, Taf. 12). To the same artistic tradition belong a horse-bridle made of deer antler, found in LH I levels at the settlement of Mitrou, East Locris (Maran & van de Moortel 2014), round ‘buttons’ from the citadel of Mycenae (Schliemann 1878: 84-86, 89), pommels and gold plates of sword handles from Grave Circle A (Karo 1930-33: pl. LXXV) and from the rich LH IIA grave of Staphylos at Skopelos (Platon 1949: pls la, Ib, II, fig. 1).
2These artefacts have been considered as evidence of far-reaching connections in the mid-2 “d millennium BC (e.g. Karo 1930-33: 272; Harding 1984: 189-200). Objects of this chronological horizon, made of bone or antler and decorated with wave band motifs (the so-called ‘ Wellenbandomamentik’ are encountered in a wide area from the Carpatho-Danubian region (Maran & van de Moortel 2014: 530, n. 1) to the Amuq Plain (Southern Turkey), and north of the Black Sea; they include mostly cheek-pieces of horse harness, discs, and cylindrical objects of uncertain use (Penner 1998; David 2007: 412-413). Similar motifs are also known from gold ornaments, small vessels, bronze tools and weapons from the same areas (Harding 1984: 189-200).
3Early Mycenaean examples are more closely related to the north; the examination of the horse bridle from Mitrou has shown clear typological affinities with the Carpathian basin and the lower Danube (Maran & van de Moortel 2014: 540-541); a northern link is also supported by the fact that the most common type of object with wave band decoration encountered in the Aegean, i.e. the disc-shaped ‘button’, is known almost exclusively from the Carpatho-Danubian region and the Peloponnese (David 2007: 412).
4However, the objects from Grave Circle A have several idiosyncratic features. Bone and antler pieces from other areas are not covered with gold, whereas accessories of rhomboid shape are so far unique to Mycenae – although they share common decoration with the disc-shaped items (i.e. the ‘moderately arched wave meander’ style, David 2007: 412) and belong clearly to the same artistic tradition.
5Those remarks, combined with the extremely limited geographical (Mycenae, Kakovatos, Mitrou and Skopelos) and temporal distribution of such items in Greece (found mostly in LH I-IIA contexts), pose questions about the nature of relations between Early Mycenaean elites and the Carpathian basin. Extensive stylistic analysis (Penner 1998; David 2001) has not identified yet the precise links between the two areas. It is still unclear whether we should speak of importation, plain imitation of decorative styles, or a case of technological transfer – and in which direction (see Maran & van de Moortel 2014, 543-545).
6Perhaps some clues towards the unraveling of the mystery can be provided by examining the technology of those items. Robert Laffineur has long stressed the importance of differentiating between networks of exchange of material goods and networks of exchange of knowledge and information (1990-91: 246-248). More recently, Romain Prévalet, in his study of Bronze Age Mediterranean jewellery (Laffineur 2013: 331-336), has made a perceptive distinction between stylistic and technical affinities, and suggested that the former may be based on mere imitation, while the latter depend on sustained training, personal contact and apprenticeship. Prévalet sees crafting techniques as chains of movements and gestures which have been repeated so many times by the craftsman, that they occur almost automatically. For that reason, he argues, we should first try to establish the stages of production (the chaîne opératoire) for each technique before attempting to compare between different regions and periods.
7Bernard Hansel has followed such an approach in his study of an antler plaque with wave band decoration from Feudvar in Northern Serbia (Hansel 1997). He has examined the way spirals were created with compasses and freehand engraving, using a complex system of measurements, and suggested that the local artisan tried to imitate the much more perfectly executed designs seen on ornaments from Mycenae (Hansel 1997: 255).
8In this paper, we will present the results of a detailed technological study of rhomboid accessories from the Mycenae Shaft Graves. We have chosen these objects because they combine a shape unique to Mycenae with the widespread ‘wave band decoration’; for that reason, they are likely to have been manufactured locally but under foreign inspiration. After discussing their archaeological context, we will outline the most important findings from their macroscopic and microscopic examination. Then, we will present the results of an experimental reconstruction of a rhomboid ‘button’, and make tentative suggestions about the tools used and the stages of manufacture. Apart from the fact that a step-by-step reconstruction of such items has never been attempted in the past, it is worth noticing that the experiment was conducted by a professional goldsmith, Mr Akis Goumas, with long experience in the study of ancient gold artifacts and the identification of tool traces on them. The experimental reconstruction allowed us to shed light on many manufacturing details, and also brought to the fore new questions that could not have been addressed otherwise.
1. The archaeological context
9Twelve rhomboid accessories made of bone and gold were recovered from Grave IV (Schliemann 1880: 260- 261; Karo 1930-33: 89-91 and pl. LXI) and another eight from Grave V (Schliemann 1880: 325, 327; Karo 1930-33: 127 and pl. LXVI). Some of them were found in pairs (nos. 340, 341, 668-670, Karo 1930-33: 90, 127). In size, they range from 3.8 x 2.8 cm to 9.1 x 5.6 cm. According to the notebook of Panayotis Stamatakis, supervisor of Schliemann’s excavations on behalf of the Athens Archaeological Society, they all seemed to have furnished male burials (Stamatakis’ notebook is part of the NAM archive). This has been confirmed by the most recent anthropological study (Papazoglou et al. 2010). Their association with male burials is reinforced by the fact that Graves IV and V also produced bronze swords with elaborate gold-plated handles, decorated with the same patterns as the button-like objects (Poursat 2014: 36-37).
10According to anthropological data (Papazoglou et al. 2010: 161-3), Grave IV held the remains of three men and possibly two women. Six rhomboid accessories furnished burial Ξ (MYC2, IV), a man 25 to 35 years old, who was also provided with the silver bull’s head rhyton, the gold lion’s head rhyton, and a gold band wrapped around an arm bone; another six rhomboid accessories accompanied male burial O (MYC3, IV), which was associated with the large silver krater with the battle scene. Grave V held the remains of four men and possibly one woman; six rhomboid accessories were found close to male burial T (who also bore the so-called ‘Agamemnon mask’, the ornate gold breastplate, the eagle pendants, and gilded bronze weapons), and another two inside the delicate alabaster vessel – along with dozens of round ‘buttons’ – in the space between male burials Y and ɸ, furnished mostly with bronze weapons and vessels. Among the finds were also three bone tubes and a flat, bone spatula-like piece with similar decoration (Schliemann 1880: 328-329).
11All rhomboid accessories consisted of a flat bone core with a torus base (Fig. 15.1). The upper surface of the core was carved with curvilinear and rectilinear motifs (in the ‘wave band style’) and covered with gold foil, wrapped around the edges. Only a few of the them are preserved intact; most have a worn bone surface, while from others survives only the gold foil that originally covered them. The intact ones range in thickness between 0.6 and 1.2 cm.

Fig. 15.1. Rhomboid accessory NAM 346: the gold leaf, the worn bone core and a section of the core with the torus base (photo: National Archaeological Museum, Athens, Greece; drawing: Akis Goumas)
12These items are usually interpreted as ‘buttons’ or clothing ornaments; this is partly due to their brooch-like shape, and partly due to the low torus base wich would ensure their attachment onto a garment or a shroud by means of a thread (Konstantinidi 2014: 147). However, since bone and antler objects with wave band decoration in other areas are frequently associated with the use of horses (Penner 1998: 160; David 2007: 412), the interpretation of Mycenaean ‘buttons’ may deserve further consideration (although no remains of horses are known from Grave Circle A, and the only horse-related finds are four possible bone cheek pieces – see Penner 1998: pl. 1: 3-6 – the iconography of the grave stelae, and a gold signet ring from Shaft Grave IV with a two-horse chariot scene on the bezel.

Fig. 15.2. Rhomboid accessory NAM 669A, front and back side (photo: National Archaeological Museum, Athens, Greece)

Fig. 15.3. Rhomboid accessory NAM 669B, front and back side (photo: National Archaeological Museum, Athens, Greece)
13In any case, the high status of the male burials to which those items are attributed suggests that they were perceived as emblems of male nobility – like so many objects in Grave Circle A (Voutsaki 2004: 358-361). This is also supported by their unique shape and the embellishment of bone with gold, which is so far unparalleled outside Mycenae. The practice of coating bone with gold foil in order to add a ‘luxurious’ tone is a feature only met in Grave Circle A and may be considered as the ‘trademark’ of a specific Aegean workshop (for the local transformation of imported or imitated ‘waveband’ objects, see Maran & van de Moortel 2014: 533-535, 540-545).
14It is also important to stress that, while in the Aegean such objects occur primarily in high-status funerary contexts (with the significant exception of Mitrou), in the Danubian region and Western Asia they are almost exclusively found in settlements, indicating that they may have not been considered so prestigious or that they were used by wider social or cultural groups (David 2007: 412, 415). In Greece, the ‘wave band tradition’ seems to have been a short-lived fashion or code, adopted by Early Mycenaean elites as a major symbol of status (Maran & van de Moortel 2014: 544), either through imitation by a local workshop or in the frame of gift exchange. But as was the case with many ‘fashions’ seen in the heterogeneous assemblage of Grave Circle A, it did not outlive its era and most probably died out after LH IIA.
2. The macroscopic and microscopic examination
15For the purposes of this study, we examined macroscopically several rhomboid ‘buttons’ from Grave Circle A, and microscopically (with a Leica DMS 300 digital microscope and a portable Dino Lite digital microscope) the following two:
16a) NAM 669A (one of two almost identical pieces from Grave IV, listed as NAM 669, see Figs 15.2-15.3), preserved in excellent condition (with the gold sheet firmly attached on the bone core) (Fig. 15.2);
17b) NAM 346 from Grave V, in which the gold sheet has been detached from the bone, thus allowing for observations and measurements on the core and the underside of the gold sheet (Fig. 15.1). The latter was further analysed with a portable XRF by Dr Gianis Bassiakos and Dr Liana Philippaki of the Demokritos Research Centre. Here are the most important remarks:
2.1. Metrics
NAM 669A: | |
Dimensions: | 7.5 x 4.6 cm, thickness 1 cm. |
NAM 346: | ||
Dimensions: | 7.8 x 5 cm, thickness | 1 cm. |
Thickness of gold sheet: | 0.04-0.06 mm. | |
Depth of bone curving: | 0.35-0.4 mm. | |
Width of bone incisions: | 0.3-0.4 mm. |
18Composition of gold foil (p-XRF analysis at 40kV, 120μA for 150sec):
Front side: | Back side: |
Au 80.87 % | Au 84.05 % |
Ag 18.38 % | Ag 15.42 % |
Cu0.71 % | Cu 0.50 % |
Fe 0.04 % | Fe 0.03 % |
19(According to Dr Bassiakos, the small differences in the composition of the two sides are analytically insignificant, suggesting the use of a single native alloy).
2.2. Traces of tools (direct and indirect)
2.2.1. Cut-marks
20No visible cut-marks are preserved on the heavily worn bone surface of NAM 346 or the eroded back side of NAM 669A.
2.2.2. On circular motifs
21The use of a standard compass for drawing circles on such items has been widely assumed because of the dots left by the pointed end of the shaft, and the perfection of execution (Karo 1930-33: 264-272; Harding 1984: 198- 199; Hansel 1997). A standard compass, however, would have produced a slanting profile on the sides of the incisions. Microscopic views (Figs 15.4-15.6) show clearly that all circular incisions have vertical (or very close to vertical) sides, suggesting the use of a different rotating device (see section 3.2.3).

Fig. 15.4. Microscopic views of the gold surface of NAM 669A (left) and the bone core of NAM 346 (right) (photos by the authors)
22Small circles used as filling ornaments present irregularities (Fig. 15.5 left), which suggest that they were initially drawn with a rotating device but then finished freehand. Similar irregularities can be seen at the curving ends of spirals (Fig. 15.5 right), suggesting that they were also made freehand.

Fig. 15.5. Microscopic views of NAM 669A, showing irregularities in the outline of circular motifs (left) and in the curving ends of spirals (right) (photos by the authors)
2.2.3. On straight lines
23Microscopic views show concurrent irregularities and imperfections (not visible to the naked eye) in the incised lines which connect confronting arcs (Fig. 15.6), indicating that they were made freehand (see also Hansel 1997: 254). The incisions have vertical sides suggesting the use of a pointed graver of rectangular section.

Fig. 15.6. Microscopic views of the gold surface of NAM 669A (left) and the underside of NAM 346 (right), showing irregularities in the lines which connect confronting arcs (photos by the authors)
2.2.4. On the background
24The background lies at a depth of 0.35-0.4 mm from the surface of motifs. All carvings have very sharp edges both on their upper and lower ends (Figs 15.4-15.6). This suggests the use of a pointed tool for sharpening the edges of motifs after excess bone material had been removed.
2.2.5. On the edges of the gold sheet
25Microscopic views of the underside of the ‘buttons’ (Fig. 15.7) suggest the use of a sharp flat chisel for cutting the sheet in the desired form and size. The foil was then sliced in the points where the core formed angles in order to avoid excessive ‘folding’.

Fig. 15.7. Microscopic views of the back side of NAM 346 (left) and NAM 669A (right), showing the folding of the gold sheet (photos by the authors)
2.2.6. On the surface of gold
26Dressed carvings have very sharp upper and lower edges (Figs 15.4-15.6), suggesting that the gold sheet was carefully stretched into the incised slots with a pointed tool. Such a tool should have been made of bone, for a bronze one could easily tear the gold foil. The shallow parallel lines seen on the surface of NAM 669A (Fig. 15.8) may have been the result of the final polishing process, most probably with a flat bone tool.

Fig. 15.8. Microscopic view of NAM 669A showing shallow parallel lines on the gold surface, possibly caused by the final polishing process (photo by the authors)
3. The experimental reconstruction
27Based on the above remarks, Mr Akis Goumas attempted to reconstruct the rhomboid accessory NAM 669A (Fig. 15.2). The object bears incised decoration of two confronting coiled bands terminating in free curved ends, and has integral double or triple discs at the angles. In the comers of the lozenge there are tiny concentric circles.
3.1. Materials and tools used in the experiment
3.1.1. Gold
28The foil used was of refined gold (99 %). This made it softer and more malleable than the original, which according to the p-XRF conducted in NAM 346, was a natural admixture of gold and silver (with tiny amounts of copper and iron). The foil was hammered to a thickness of 0.05 mm, very much like the original.
3.1.2. Bone
29According to the zooarchaeologist Dr Dimitra Mylona, who made a preliminary macroscopic examination of the rhomboid ‘buttons’, most cores were probably made of different bones of large mammals (long bones, flat bones – i.e. scapula and pelvis – and probably ribs); however, the morphology of some examples suggests that they may have been produced from deer antler (we are grateful to Dr Mylona for allowing us to mention this information). A detailed study of the cores will be undertaken soon. For this experiment, the long bone of a cattle was used. The bone was first boiled and then soaked in sodium hypochlorite solution to clean off remains of flesh and grease.
3.1.3. Tools
30All tools employed in the experiment were made by Akis Goumas of materials used in the LB A (i.e. wood, bone, bronze and stone) (Fig. 15.9). The tools are described in detail below.

Fig. 15.9. The tools used in the experiment (photo by the authors)
3.2. Description of the procedure
3.2.1. Cutting the bone
31The usable piece was cut off the long cattle bone with a bronze saw, and trimmed in the comers to obtain a hexagonal shape (Fig. 15.10). It was then polished with a grindstone to erase cut-marks.

Fig. 15.10. The cutting of the bone with a bronze saw (photos by the authors)
3.2.2. Design composition and transfer onto the bone
32The symmetry of design (see Figs 15.2-15.3) suggests a precise system of measurements based on some sort of grid (Karo 1930-33: 264; Hansel 1997: 253-254 and fig. 2), like that known from other sectors of art in that period (Shaw 2003). Such a grid could not have been drawn directly on the bone for it would leave marks on the surface. Mr Goumas drew the basic design (six ¾ circles corresponding to the main arcs of the ornament) on a flat piece of baked clay with the help of a plain grid (Fig. 15.11 left, see also Appendix and Fig. 15.27 for a detailed description of the drawing process). Then he pressed a thin bronze sheet on the clay plaque and used a pointed tool to pierce the central dots of the six circles (Fig. 15.11 middle). By placing the bronze sheet on the bone, he used the piercings as a guide to mark the surface (Fig. 15.11 right). These marks provided stable reference points for the carving process.

Fig. 15.11. The transfer of the basic design from the clay plaque to the bone surface with the help of a bronze sheet (photos by the authors)
3.2.3. Carving the main motifs
33For reasons outlined above (see section 2.2.2), the use of a standard compass for drawing circles on the bone surface should be excluded. The observed morphology of incisions (sharp vertical sides) suggests the use of a rotating device with a carving end that applied vertically on the bone surface. For that reason, Akis Goumas chose to create a number of bronze gravers with parallel projecting points at varying distances (Fig. 15.12). These double-pointed gravers were used for carving ¾ circles, which could then be connected freehand to form continuous spirals. Although we have not managed to identify such tools in LBA contexts, recently it was brought to our attention – by Dr Despina Ignatiadou, Head of the Sculpture Collection at the NAM, to whom we are grateful – that a similar item has been excavated at an Archaic grave in Tsaousitsa, Macedonia (Adam-Veleni & Koukouvou 2012: 165).

Fig. 15.12. The carving of the bone core with rotating devices featuring two parallel projecting points (photos by the authors)
34As suggested earlier (see section 2.2.3), the lines connecting confronting arcs were carved freehand. A pointed bronze graver of rectangular section was used for this purpose (Fig. 15.13). Mr Goumas filled the incisions with charcoal to make them more clearly visible during the carving process.

Fig. 15.13. The carving of lines connecting confronting arcs with a pointed tool (photos by the authors)
35For longer lines, a sturdier device that could cut the bone deep and straight was necessary. In the experiment, we used a long graver with indentations, provided with a wooden handle that enabled the craftsman to control movement and not deviate significantly from the straight line (Fig. 15.14).

Fig. 15.14. The carving of long straight lines using a graver with indentations (photos by the authors)
3.2.4. Finishing decoration
36The ends of spirals were formed by connecting the parallel contours of concentric circles. This was made freehand (Fig. 15.15 left) both because of the small available space and because the spiral ends of the original ‘buttons’ present irregularities, which do not conform to the use of a compass or other rotating device (Figs 15.5, 15.6). Additional circles were then carved in the free spaces. The decoration of the external discs was made with rotating double-pointed gravers. After drawing the motifs, the rest of the surface had to be carved to a depth of 0.4 mm to create the background. Mr Goumas used a flat toothed chisel to remove excess material and a pointed tool to sharpen the edges (Fig. 15.15 right).

Fig. 15.15. The carving of spiral ends (left) and the sharpening of the background (right) with a pointed tool (photos by the authors)
3.2.5. Cutting excess bone and polishing
37Once the carving process was completed, the bone was cut into its final shape. Since no cut-marks were visible on the bone core of NAM 346, we could not be sure about the details of the technique. In the experiment, Mr Goumas sliced the pieces between the projecting discs into vertical stripes with a bronze saw (Fig. 15.16 left), and then broke and removed the stripes with a flat bronze tool. The surface was then smoothened with a grindstone (Fig. 15.16 right), and polished with a wooden tool covered with leather impregnated in a mixture of animal glue and emery powder.

Fig. 15.16. Cutting excess material with a bronze saw (left) and smoothening the surface with a grindstone (right) (photos by the authors)
3.2.6. Cutting the gold sheet
38The hammering of the gold leaf into a thickness of 0.05 mm lasted for a quite long time and necessitated the frequent re-heating of the sheet. Then a pointed bronze graver was used to score the outline of the ‘button’ on the gold surface (Fig. 15.17 left) and a bronze chisel to cut the sheet (Fig. 15.17 right).

Fig. 15.17. Scoring the gold sheet with a bronze graver (left) and cutting excess material with a bronze chisel (right) (photos by the authors)
3.2.7. Folding the gold leaf around the core
39In order to ‘read’ the impressed motifs of the bone core, the gold leaf has to be tightly stretched and wrapped around the core. For that reason, Mr Goumas created a clay ‘depression’, in which he could firmly press the bone upon the gold leaf and wrap the latter around the former (Fig. 15.18 left). Before wrapping, he sliced the gold leaf at the points where the core formed angles, to avoid excessive folding (Fig. 15.18 middle). Then, he wrapped the gold around the core and used a flat bone tool to smoothen the foldings (Fig. 15.18 right).

Fig. 15.18. Placing the ornament in a clay depression and wrapping the gold sheet around the bone core (photos by the authors)
3.2.8. Impressing the gold leaf upon the carved bone core
40This process was executed in two stages: first, the outline of decorative motifs was impressed, and then the gold leaf was stretched into the deep carvings. As microscopic views show (Figs 15.4-15.6), all dressed carvings have very sharp edges, suggesting the use of a pointed tool. In this experiment, we used a bone tool with one flat and one pointed end. The flat end allowed the gold to ‘read’ the outline of the decoration (Fig. 15.19 left), while the pointed end stretched the metal into the carvings to produce sharp edges (Fig. 15.19 middle). Finally, the flat end of the tool was used again to polish the surface (Fig. 15.19 right). Such double-ended bone tools are known from the palatial workshops of Mycenae, dated to the 13th c. BC (i.e. NAM Inv. no. P 1060).

Fig. 15.19. The impression process: ’reading’ the motifs (left), stretching the gold into the slots (middle) and polishing (right) (photos by the authors)
4. Comparative examination
41Following the experiment, the reconstructed ‘button’ was examined microscopically against NAM 669A. Although the quality of craftsmanship is certainly inferior to the original, the manufacturing details show impressive similarities:
421. Incisions are equally deep and have similar vertical sides. Moreover, the reconstructed ‘button’ shows the same horizontal lines caused by the final polishing as on the original (Fig. 15.20).

Fig. 15.20. Microscopic views of the central part of the gold surface in the modern replica (left) and on NAM 669A (right) (photos by the authors)
432. The straight lines connecting circular motifs in the replica present the same irregularities as on the original (Fig. 15.21). The same is true for curving lines connecting the ends of parallel contours of circles (Fig. 15.22).

Fig. 15.21. Microscopic views showing irregularities in the lines which connect circular motifs on the modern replica (left) and on NAM 669A (right) (photos by the authors)

Fig. 15.22. Microscopic views showing irregularities in the curving lines which connect parallel contours of circles on the modern replica (left) and on NAM 669A (right) (photos by the authors)
443. Smaller circles present the same irregular outline as on the original (Fig. 15.23), suggesting that they were initially made with a rotating device and then finished by hand.

Fig. 15.23. Microscopic views showing irregularities in the outline of circular motifs on the modern replica (left) and on NAM 669A (right) (photos by the authors)
454. The outlines of the lozenges are equally straight and sharp (Fig. 15.24).

Fig. 15.24. Microscopic views showing the morphology of the long straight lines of the lozenge on the modern replica (left) and on NAM 669A (right) (photos by the authors)
465. In general, the original has sharper lower edges, suggesting that the Mycenaean craftsman was more efficient either in carving the background or in stretching the gold leaf into the carved spaces, which may suggest the use of finer tools (Figs 15.20-15.23).
476. The foldings around the edges of the bone core present striking similarities (Fig. 15.25).

Fig. 15.25. Microscopic views of the underside of the modern replica (left) and of NAM 669A (right) (photos by the authors)
Concluding remarks
48The attempted experiment was based on detailed macroscopic and microscopic observations and provided valuable information about the tools and techniques used in the manufacture of rhomboid ‘buttons’. The comparative examination of the final product against the original showed striking similarities. This suggests that, even if we cannot be sure about the exact form of the Mycenaean tools, their overall morphology cannot have differed significantly from the ones used in the experiment. What it more, the experiment has established in a reliable way the sequence of movements and actions performed by the craftsman, which is summarised in Fig. 15.26.
49The most demanding task of the process was the carving of the bone. In absolute terms, it required approximately 75- 80 % of the total manufacturing time. In that sense, Mycenaean ‘buttons’ are correctly treated as part of the widespread tradition of bone and antler objects with wave band decoration rather than as ‘gold jewellery’, and should be compared with their Carpatho-Danubian counterparts in terms of technique.
50The examined objects demonstrate a high level of expertise. Harding has suggested that experimental or low-skill items of that type are lacking from Greece and concluded that “one might more logically suggest that the Mycenaean art motifs came directly from the Carpathian Basin, since they appear to arrive fully developed” (1984: 199). Karo, on the other hand, has suggested possible evolutionary stages of this technique at Mycenae (1930-33: 268-272). What we have learned from this experiment is that an experienced craftsmen could have quickly acquired the necessary skills to produce pieces of very high quality (although the carving of circular motifs on cylindrical objects, such as the horse bridle from Mitrou, may have necessitated different techniques and tools).
51NAM 669A is one of the most technically advanced ‘buttons’ at Mycenae. Further study may allow us to test whether Karo’s identification of evolutionary stages can be confirmed technologically. If so, one could confidently speak of a local workshop. As to questions of technological transfer, our study seems to confirm Hansel’s suggestion that the Mycenaean accessories were much more accurately designed than the Feudvar antler piece (Hansel 1997: 255). However, more analyses of this type (based on microscopic study and experimentation) are necessary before drawing sound conclusions. As Joseph Maran has rightly pointed out (pers. comm.), it is probable that the exchange of technical knowledge was a multi-directional process, including an initial stage, during which the idea of decorating antler and bone pieces with wave band motifs came to Greece from the north, and a later one, during which improved methods of decoration that developed at Mycenae influenced artisans in the Carpathian basin (see also Maran & van de Moortel 2014: 544-545). We hope that the methodology suggested here opens a path to the direction of such type of research.
52The replica of NAM 669A was made for research purposes, but also in order to be presented at the exhibition ‘The Greeks. From Agamemnon to Alexander the Great’ (Canada, USA 2015-2016). For a few months, the two ornaments (the original and its replica) were displayed side by side in the same case, along with the tools used at the experiment, so that the visitors could appreciate the challenges faced by the Mycenaean craftsman and the inventiveness with which he managed to overpass them. The attempt to reproduce such a masterpiece was not a mere research enterprise; rather, it was an invitation from a fellow craftsman to another on a journey in time: the anonymous Mycenaean artist and Akis Goumas could never have met otherwise – separated as they are by 36 centuries – but only through their work, mentally combining their forces to recreate a scene from a NE Peloponnesian workshop at the very beginning of the Late Bronze Age.

Fig. 15.26. Suggested stages of manufacture for a rhomboid accessory (drawings by Akis Goumas)
Bibliographie
References
▪ Adam-Veleni & Koukouvou 2012 = P. Adam-Veleni & A. Koukouvou, Archaeology behind battle lines, in Thessaloniki of the Turbulent Years 1912-1922, Exhibition Catalogue, Archaeological Museum ofThessaloniki, edited by P. Adam-Veleni & A. Koukouvou, Thessaloniki (2012), 165.
▪ David 2001 = W. David, Zu den Beziehungen zwischen Donau-Karpatenraum, osteuropäischen Steppengebieten und ägäisch-anatolischem Raum zur Zeit der mykenischen Schachtgräber unter Berücksichtigung neuerer Funde aus Südbayern, in The Mediterranean and Central Europe in Contacts and Confrontations. From the Bronze Age, edited by M. Novotná, Trnava (2001), 51-80.
▪ David 2007 = W. David, Gold and bone artefacts as evidence of mutual contact between the Aegean, the Carpathian Basin and southern Germany in the second millennium BC, in Between the Aegean and Baltic Seas. Prehistory Across Borders, edited by I. Galanaki, H. Tomas, Y. Galanakis & R. Laffineur (Aegaeum 27), Liège (2007), 411-420.
▪ Dickinson 1977 = O.T.P.K. Dickinson, The Origins of Mycenaean Civilization, (SIMA 49). Göteborg (1977).
▪ Harding 1984 = A. F. Harding, The Mycenaens and Europe, London (1984).
▪ Hansel 1997 = B. Hansel, Die Quadratur des Kreises in den Bronzezeit Serbiens, in ANTIΔΩPON. Dragoslavo Srejović completis LXVannis ab amicis collegis discipulis oblatum, edited by M. Lazić, Belgrade (1997), 251- 258.
▪ Hiller 1991 = S. Hiller, The Mycenaeans and the Black Sea, in THALASSA. L’Egée préhistorique et la mer, edited by R. Laffineur & L Basch (Aegaeum 7), Liège (1991), 207-216.
▪ Karo 1930-33 = G. Karo, Die Schachgräbervon Mykenai, Lund (1930-33).
▪ Konstantinidi 2014 = E. Konstantinidi-Syvridi, Buttons, pins, clips and belts..., ’Inconspicuous’ dress accessories from the burial context of the Mycenaean period (16th to 12th cent. BC), in Prehistoric, Ancient Near Eastern and Aegean Textiles and Dress, an Interdisciplinary Anthology, edited by M. Harlow, C. Michel & M.-L. Nosch, Oxford & Philadelphia (2014), 143-157.
▪ Laffineur 1990-91 = R. Laffineur, Material and craftsmanship in the Mycenae Shaft Graves: imports vs local production, Minos 25-26 (1990-91), 245-295.
▪ Maran & van de Moortel 2014 = J. Maran & A. van de Moortel, A horse-bridle piece with Carpatho-Danubian connections from Late Helladic I Mitrou and the emergence of a warlike elite in Greece during the Shaft Grave period, AJA 118 (2014), 529-548.
▪ Midler 1909= K. Müller, Alt Pylos. II. Die Funde aus den Kuppelgräbern von Kakovatos, AM 34 (1909), 269- 328.
▪ Papazoglou-Manioudaki et al. 2010 = L. Papazoglou-Manioudaki, A. Nafplioti, J.H. Musgrave & A.J.N.W. Prag, Mycenae revisited, Part 3. The human remains from Grave Circle A at Mycenae. Behind the masks: a study of the bones of Shaft Graves I-VI, BSA 105 (2010), 157-224.
▪ Penner 1998 = S. Penner, Schliemanns Schachtgraberrund und der europäische Nordosten. Studien zur Herkunft der frümykenischen Streitwagenausstattung, Bonn (1998).
▪ Platon 1949 = N. Platon, O τάφος του Σταφύλου και ο μινωικός αποικισμός τη Πεπαρήθου, Kretika Chronika 3 (1949), 534-573.
▪ Poursat 2014 = J.-C. Poursat, L’art égéen, 2. Mycènes et le monde mycenien, Paris (2014).
▪ Prévalet 2013 = R. Pre valet, La decoration des pièces d’orfèvrerie-bijouterie en Méditerranée orientale à l’âge du Bronze: techniques, productions, transmissions, PhD thesis, Université Paris 1, Panthéon-Sorbonne (2013).
▪ Shaw 2003 = M.C. Shaw, Grids and other drafting devices in Minoan and other Aegean wall painting. A comparative analysis including Egypt, in METRON. Measuring the Aegean Bronze Age, edited by K. P. Foster & R. Laffineur (Aegaeum 24), Liège (2003), 179-189.
▪ Schliemann 1880 = H. Schliemann, Mycenae. A Narrative of Researches and Discoveries at Mycenae and Tiryns, New York (1880).
▪ Schliemann 1878 = H. Schliemann, Mykenae. Leipzig (1878).
▪ Vermeule 1975 = E. Vermeule, The Art of the Shaft Graves of Mycenae, Cincinnati (1975).
▪ Vbutsaki 2004 = S. Vbutsaki, Age and gender in the southern Greek Mainland, 2000-1500 BC, Ethnographisch-Archäologisches Zeitschrift 45 (2004), 339-363.
Annexe
Appendix
The principles of the design
To achieve symmetry of design on rhomboid accessories, the craftsman had to use basic geometric principles. This has been noted by various authors, who have attempted to reconstruct the process (Karo 1930-33: 264-269; Harding 1984: 198-199; Hansel 1997). Here, we propose a refined version of the drawing method, which is based on the results of our experiment and requires only the use of a plain compass and a ruler (Fig. 15.27).
a. With the help of a compass set at a standard radius (r), the craftsman divides a line into 8 equal parts. Then he draws two circles of radius (r) centered on points A and B.
b. To produce vertical reference lines, the craftsman draws arcs with radius > (r) from points C, D and E. By connecting the crossing arcs, he draws vertical lines. Then, he connects the points where the vertical lines cross the circles, to draw the horizontal lines of a square F-G-H-I.
c. Using the corners of the square F-G-H-I as centres, the craftsman draws four tangent circles with radius (r). The tangent points and point D define a central vertical line
d. To draw the outline of the lozenge, the craftsman links the outer dots of the horizontal line (J and K) with the perimeter of the corresponding circles and the central vertical line.
e. To fill the empty space on either side of the lozenge, the craftsman draws two circles centred at points C and E with radius slightly larger than (r).
f. When measurements are transferred onto the bone surface (Fig. 15.11), the craftsman carves six ¾ concentric circles and then connects their edges by hand.

Fig. 15.27. The motif drawing process (drawings by Akis Goumas)
Notes de bas de page
1 Professor R. Laffineur is one of the most meticulous researchers of the Aegean Bronze Age. His studies have been crucial, among others, for helping us to appreciate the cultural and social significance of Early Mycenaean jewellery. This paper is a small contribution to honour his retirement after decades of hard and productive work.
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Du Phénotype au génotype
Analyse de la syntaxe spatiale en architecture minoenne (MMIIIB – MRIB)
Quentin Letesson
2009
PERIFEREIA Étude régionale de la Crète aux Minoen Récent II-IIIB (1450-1200 av. J.-C.)
La Crète centrale et occidentale
Charlotte Langohr
2009
Minoan Realities
Approaches to Images, Architecture, and Society in the Aegean Bronze Age
Diamantis Panagiotopoulos et Ute Günkel-Maschek (dir.)
2012
Excavations at Sissi III
Preliminary Report on the 2011 Campaign
Jan Driessen, Charlotte Langohr, Quentin Letesson et al.
2012
Minoan Archaeology
Perspectives for the 21st Century
Sarah Cappel, Ute Günkel-Maschek et Diamantis Panagiotopoulos (dir.)
2015
THRAVSMA
Contextualising the Intentional Destruction of Objects in the Bronze Age Aegean and Cyprus
Kate Harrell et Jan Driessen (dir.)
2015
How Long is a Century?
Late Minoan IIIB Pottery. Relative Chronology and Regional Differences
Charlotte Langohr (dir.)
2017