Late Bronze Age Glass Production on Rhodes, Greece

REPRINTED FROM

JOURNAL OF GLASS STUDIES

VOLUME 54 • 2012

Copyright © 2012 by The Corning Museum of Glass, Corning, NY 14830-2253

Pavlos Triantafyllidis and Ioannis Karatasios

Late Bronze Age Glass Production on Rhodes, Greece

25

settlement of Trianda near Ialysos, offer abun-dant evidence of the important role that Rhodes played as a commercial partner. Nearly 3,000 glass artifacts have been found at these two cem eteries, and also at cemeteries in southern Rhodes,3 including those at Pylona,4 Passia, Le-los, Apsaktiras, Apolakkia, Gennadi, and Lin-dos. Most of these items were used as toiletries and for personal adornment, but a few were core-formed vessels that were imported to Rhodes from the thriving glass workshops of Pharaonic Egypt between the 14th and 12th cen-turies B.C.5

This body of material includes both mono-chrome and multicolored glass beads in a variety of shapes and patterns. The most common cat-egory is cast blue or translucent turquoise relief beads that were used in necklaces and sewn, ei-ther singly or in groups, on garments. Four such beads, decorated with spiral motifs imitating curls of hair, were employed in diadems, sym-bols of wealth and elevated social status. Two

ThE IsLANd of Rhodes is widely regard-ed to have been one of the most impor-tant eastern centers of the Minoan and

Mycenaean cultures,1 where trade involving the Aegean, the wider Mediterranean basin, and the Near East flourished from the LB I (1600–1500 B.C.) to Lh III C (1190–1070 B.C.) periods.

during the Mycenaean era, Rhodes was an important link between the East and the West. The island had developed both a cosmopolitan character and strong relations with neighbor-ing areas, especially Crete, Cyprus, Egypt, and northern syria. All sorts of goods and raw mate rials that traveled along sea routes were exchanged via Rhodes. The island also came in con tact with new technological advances from the East, including those pertaining to such vit-reous materials as faience and glass, during the Late Bronze Age, particularly from the Lh III A 1 to Lh III C periods (1390–1070 B.C.).2

The rich Mycenaean cemeteries of Ialysos and Kamiros, as well as the major prehistoric

Late Bronze Age Glass Production on Rhodes, Greece

Pavlos Triantafyllidis and Ioannis Karatasios

Acknowledgments. We thank Toula Marketou, archaeologist in the 22nd Archaeological Ephorate of Prehistoric and Clas-sical Antiquities, and excavator of the settlement at Trianda, Ialysos, on Rhodes, for permission to publish the LB glass cul-let and to conduct the chemical analysis in the National Center for scientific Research “demokritos” in Athens.

1. Toula Marketou, “Rhodes,” in The Oxford Handbook of the Bronze Age Aegean, ed. Eric h. Cline, Oxford: Oxford University Press, 2010, pp. 775–793; idem, “Vulkanische und Überschwemmungen an der Ialysos-siedlung in der spätbronze-teit,” Archäologische Forschung und Funde in der Dodekanes: Rhodos, Ialyssos, Kos, Nisyros und Giali, Weilheim: Verein zur Förderung der Aufarbeitung der hellenischen Geschichte, 2007, pp. 172–180.

2. Mario Benzi, Rodi e la Civiltà Micenea, Rome: Gruppo Editoriale Internazionale, 1992, pp. 195–201; Pavlos Trianta-fyllidis, “Evidence and Glass Workshops in Rhodes from Pre-historical until the Early Christian Periods,” in The Glass from

Antiquity until Today, ed. Petros Themelis, Athens: society of Mes senian Archaeological studies, 2002, pp. 4–55, esp. pp. 41–43 (in Greek; French summary, p. 55); idem, “The Art and the Craft of Glass in south-East Aegean,” in Hyalos = Vitrum = Glass: History, Technology and Conservation of Glass and Vitreous Materials in the Hellenic World, ed. G. Kordas, Athens: Glasnet Publications, 2002, pp. 21–40, esp. pp. 22–24.

3. Pavlos Triantafyllidis, “Glass and Faience Minor Objects from southern Rhodes,” Dodecanesiaka Chronika, v. 24, 2010, pp. 460–479 (in Greek).

4. Efi Karantzali, The Mycenaean Cemetery at Pylona on Rhodes, BAR International series, no. 988, Oxford: Archaeo-press, 2001, pp. 73–81 and 117–118.

5. Pavlos Triantafyllidis, “An Egyptian Core-Formed Krater-iskos in the Rhodes Archaeological Museum, Greece,” Journal of Glass Studies, v. 50, 2008, pp. 295–297 (six fragments of core-formed glass vessels).

26

Rhodes that preserve, in rare instances, a thin layer of gold foil on the surface.9 This has been interpreted as an attempt on the part of glass-makers to imitate precious gold artifacts. Princi-pal examples of this practice include necklaces, found at Ialysos, made of blue glass disk beads in the form of a rosette or a vase (Fig. 2); each of these beads was carefully wrapped in thin gold foil. Another group of these beads, also from Ialysos, was decorated with a curled leaf motif and a gold dot (Fig. 3). They have square ends, with inserted small gold disks, a practice

of these special beads were found in the Myce-naean cemetery at Pylona,6 and the other two were uncovered in the Mycenaean cemetery at Aghios Georgios Koutsos in Gennadi (Fig. 1).7 From their context, they have been dated be-tween the late Lh III A 2 and Lh III B 1 (1370–1200 B.C.) periods, and their design is similar to that of gold beads that were made from the 16th to early 15th centuries B.C.8

The strong connection between glassworking and goldsmithing in the 14th and 13th centu-ries B.C. can be seen in glass relief beads from

FIG. 1. Diadems of translucent blue (left) and turquoise (right) glass relief beads with spiral ornaments. Aghios Georgios Koutsos, Gennadi.

Mycenaean, late 14th–early 13th century B.C. H. 1.8 cm. Rhodes Archaeological Museum for Prehistoric Antiquities, inv. nos. Y 825

and Y 826. (Photo: Pavlos Triantafyllidis)

6. Karantzali [note 4], p. 75, fig. 42, pl. 49c, d.7. Efi Karantzali, “New sites of Mycenaean Cemeteries in

southern Rhodes,” Archaeologike Ephemeris, v. 148, 2009, pp. 223–274, esp. pp. 250–252.

8. Karantzali [note 4], p. 73; Maud spaer, Ancient Glass in the Israel Museum: Beads and Other Small Objects, Jerusalem: The Israel Museum, 2001, p. 60.

9. donald B. harden, Catalogue of Greek and Roman Glass in the British Museum, v. 1, London: British Museum Publica-tions for the Trustees of the British Museum, 1981, pp. 39–40 (rosette disks with gold foil from Ialysos); spaer [note 8]; Georg

Nightingale, “die Kombination von Gold und Glas bei my-kenischen Perlen,” Österreichische Forschungen zur Ägäischen Bronze zeit 1998, Akten der Tagung am Institut für Klassische Archäologie der Universität Wien, 2.–3. Mai 1998, ed. Fritz Bla kolmer, Vienna: Phoibos Verlag, 2000, pp. 159–165; idem, “Mykenisches Glas,” Althellenische Technologie und Technik von der prähistorischen bis zur hellenistischen Zeit mit Schwer-punkt auf der prähistorischen Epoche, 21–23.03.2003, Ohl-stadt, Weilheim: Verein zur Förderung der Aufarbeitung der hellenischen Geschichte, 2004, pp. 171–194, esp. p. 183.

27

that was also influenced by goldsmithing. A vari ation of these beads, with inlaid turquoise glass rosette disks (Fig. 4), was used as a sub-stitute for the gold disks of the beads with the curled leaf motif.

It is very likely that the same stone molds were used for the manufacture of glass and gold jewelry.10 These molds were made mainly of stea tite, which has been reported in several My-cenaean sites in Greece. One of the molds was found in a mixed and disturbed LM II / Lh III A 1 (1425–1370 B.C.) context at Trianda. It is decorated with a small papyrus form containing

FIG. 2. Necklace of cast blue glass beads and their gold covers. Rhodes, Ialysos, Tomb 4. Mycenaean, about 1400–1300 B.C. H. (beads) 1.8 cm. Trustees of The British Museum (inv. nos.

GR 1870.10-8.7 and GR 1870.10.-8.50).

FIG. 3. Three three-leaf blue glass relief beads, each with gold rosette and disk cover. Rhodes, Ialysos, Tomb 31/29. Mycenaean, 1390–1340 B.C. H. 4.9 cm. Rhodes Archaeological Museum for Prehistoric An tiq uities (inv. no. 3558). (Photo: Pavlos Triantafyl-lidis)

10. spaer [note 8], pp. 60–61; Pavlos Triantafyllidis, “Or-naments of Glass and Faience from Armenochori on Astypa-laia,” Athens Annals of Archaeology, vv. 35–38, 2002–2005, pp. 165–184, esp. pp. 177–178 (in Greek; English summary, p. 184); david F. Grose, Early Ancient Glass: Core-Formed, Rod-Formed, and Cast Vessels and Objects from the Late Bronze Age to the Early Roman Empire, 1600 B.C. to A.D. 50, New York: hudson hills Press in association with The Toledo Mu-seum of Art, 1989, p. 57; E. Marianne stern and Birgit schlick-Nolte, Early Glass of the Ancient World, 1600 B.C.–A.D. 50: Ernesto Wolf Collection, Ostfildern, Germany: Verlag Gerd hat je, 1994, pp. 49–50.

28

for the presence of secondary glass production on Rhodes during the 14th and 13th centuries B.C. Glass ingots are already known from the Uluburun shipwreck in the Aegean13 and from the eastern Mediterranean and the Near East14: Ugarit in syria, Nuzi and Tell Brak in Mesopo-tamia, Failaka (ancient dilmun) island in the

an ivy and lily hybrid motif, which is known in beads from the Mycenaean cemeteries at Ialy sos and Pylona. The mold indicates the existence on Rhodes of a Mycenaean glass workshop that was probably active during the 14th cen-tury B.C.11 Additional evidence is provided by a chunk of translucent turquoise glass (Fig. 5) that was found in a rescue excavation of the 22nd Ephorate for Prehistoric and Classical An-tiquities near the prehistoric settlement of Tri-anda. The glass chunk was found on the floor of a room that was dated to the Lh III A 1 and Lh III B 1 periods (1390–1200 B.C.)—a time during which a settlement flourished in what is known as the Markou plot at Ialysos, west of the now dried-up Trianda River.12 The glass is weathered and covered with a white crust, but it retains, in one of its surfaces, a smooth, thin layer with traces of silica, which clearly suggests that it was detached from a clay melting pot or glass crucible.

This appears to be the first solid evidence of a fragment from a chunk of raw glass found in an excavation in Greece, and it offers support

FIG. 4. Two translucent blue glass relief beads, each with inlaid turquoise glass rosette disk. Rhodes, Ialy-sos, Tomb 4/24. Mycenaean, 1390–1340 B.C. H. 4.9 cm. Rhodes Archaeological Museum for Prehistoric Antiquities (inv. no. 3521). (Photo: Pavlos Triantafyl-lidis)

FIG. 5. Fragment from chunk of turquoise glass from Trianda, Ialysos, Rhodes. Mycenaean, 1390–1340 B.C. H. 2.6 cm, W. (max). 2.1 cm. Rhodes Archaeo-logical Museum for Prehistoric Antiquities (inv. no. Y 937). (Photo: Giorgos Kassiotis)

11. Karantzali [note 4], p. 7; Triantafyllidis, “The Art” [note 2], p. 24; idem, “Evidence” [note 2], p. 42.

12. Toula Marketou, “Ialysos-Trianda: Excavation Reports,” Archaeologikon Deltion, v. 39, 1984, B, pp. 325–326; Efi Ka-rantzali, “The Mycenaeans at Ialysos: Trading station or Colo-ny?” in Emporia: Aegeans in the Central and Eastern Mediter-ranean, ed. Robert Laffineur and Emanuele Greco, Aegeaum 25, Proceedings of the 10th International Aegean Conference, Ath-ens, Italian School of Archaeology, 14–18 April 2004, Eupen: KLIEMO, 2005, pp. 141–151, esp. pp. 142–143.

13. Cemal Pulak, “das schiffswrack von Uluburun,” in Das Schiff von Uluburun, ed. Ünsal Yalçin, Cemal Pulak, and Rainer slotta, Bochum: deutsches Bergbau-Museum, 2005, pp. 55–102, and catalog, pp. 559–633.

14. Joan Aruz, Kim Benzel, and Jean M. Evans, eds., Beyond Babylon: Art, Trade, and Diplomacy in the Second Millennium B.C., New York: The Metropolitan Museum of Art, and New haven, Connecticut: Yale University Press, 2008, pp. 313–314.

29

Persian Gulf, and Tell el-Amarna and Quantir in Egypt.

Until now, the presence of Mycenaean glass workshops in Greece has been indicated mainly by the typological study of archaeological glass from Mycenae, Tiryns, and Midea in the Pelo-ponnese; from Thebes on the mainland; and from Knossos on Crete.15 however, no actual work shop installations for primary or second-ary glassmaking have been traced.

ANALYsIs OF GLAss ChUNK Y 937

The glass was cross-sectioned, vertical to the external surface, using a diamond wheel. The sample was mounted in epoxy resin, and ground and polished to 1 μm, in order to provide a flat surface for chemical analysis. Major and minor elements were determined, using an FEI Quanta Inspect d8334 scanning electron microscope (sEM) equipped with an energy-dispersive X-ray analyzer (EdX). The surface of the glass had been coated with carbon to prevent charging of the surface, and consequently any distortion and deflection of the electron beam, during the sEM examination.

The analysis was conducted at high magni-fication (up to 15,000X) by continuously scan-ning an uncorroded, inclusion-free area of the glass (5 x 5 μm), using 25 kV accelerating volt-age. Quantitative analyses were corrected with a ZAF routine. The accuracy and precision of the quantitative analysis were calculated by using Corning B as a reference glass.16

Results

The examination of samples of the glass chunk with secondary and backscattered elec-trons (Figs. 6–8) allowed three areas to be dis-tinguished and analyzed: (1) the core (Y 937-TRQ, translucent turquoise glass, Fig. 6), (2) a surface layer (Y 937-Gr, translucent green), and (3) an external yellowish layer of ceramic re-mains (Y 937-Cer). several mineral inclusions in the three layers were also analyzed. Exami-nation of the samples identified glass formers,

colorants, opacifiers, and decolorizers. The com-position of each sample area (expressed in weight percent of the oxides) is presented in Table 1.

Both Y 937-TRQ and Y 937-Gr are soda-lime-silica glasses with elevated potash (~3%) and magnesium oxide (3%–4%). These values probably indicate the use of plant ashes for in-troducing sodium into the batch.17 The use of plant ashes instead of natron in the production of Early Bronze Age and Mycenaean glasses seems to have been a common practice in Egypt and Mesopotamia.18 The concentration of cal-cium oxide (~9%) should be related to the pres-ence of antimony oxide (~3.5%) and the use of white calcium antimonate crystals (Casb2O6 or Ca2sb2O7) as an opacifier.19 The very low

15. Marina Panagiotaki, “The Technological development of Aegean Vitreous Materials in the Bronze Age,” Vitreous Ma-terials in the Late Bronze Age Aegean, ed. Caroline M. Jackson and Emma C. Wager, sheffield studies in Aegean Archaeology, v. 9, Oxford: Oxbow Books, 2008, pp. 34–63, esp. pp. 45–48.

16. Caroline M. Jackson and Paul T. Nicholson, “Composi-tional Analysis of the Vitreous Materials Found at Amarna,” in Brilliant Things for Akhenaten: The Production of Glass, Vit-reous Materials and Pottery at Amarna Site O45.1, ed. Paul T. Nicholson, London: Egypt Exploration society, 2007, pp. 101–115, esp. p. 111.

17. W. E. s. Turner, “studies in Ancient Glasses and Glass-making Processes. Part V: Raw Materials and Melting Process-es,” Journal of the Society of Glass Technology, v. 40, no. 194, 1956, pp. 277–300.

18. Christine Lilyquist and others, “Part 2. Glass,” in Stud-ies in Early Egyptian Glass, ed. Christine Lilyquist, Robert h. Brill, and Mark T. Wypyski, New York: The Metropolitan Mu-seum of Art, 1993, pp. 23–58; J. L. Mass, M. T. Wypyski, and R. E. stone, “Malkata and Lisht Glassmaking Technologies: Towards a specific Link between second Millennium BC Met-allurgists and Glassmakers,” Archaeometry, v. 44, no. 1, Feb-ruary 2002, pp. 67–82; A. J. shortland and M. s. Tite, “Raw Materials of Glass from Amarna and Implications for the Ori-gins of Egyptian Glass,” Archaeometry, v. 42, no. 1, 2000, pp. 141–151; Paul T. Nicholson and Julian henderson, “Glass,” in Ancient Egyptian Materials and Technology, ed. Paul T. Nich-olson and Ian shaw, Cambridge and New York: Cambridge University Press, 1999, pp. 195–224; Thilo Rehren, “Aspects of the Production of Cobalt-Blue Glass in Egypt,” Archaeome-try, v. 43, no. 4, November 2001, pp. 483–489; M. s. Tite and A. J. shortland, “Production Technology for Copper- and Cobalt-Blue Vitreous Materials from the New Kingdom site of Amarna: A Reappraisal,” Archaeometry, v. 45, no. 2, May 2003, pp. 285–312; Jackson and Nicholson [note 16].

19. A. J. shortland, “The Use and Origin of Antimo nate Colorants in Early Egyptian Glass,” Archaeometry, v. 44, no. 4, 2002, pp. 517–530.

30

20. Tite and shortland [note 18].21. Jackson and Nicholson [note 16]; Julian henderson,

Jane Evans, and Kalliopi Nikita, “Isotopic Evidence for the Pri-mary Production, Provenance and Trade of Late Bronze Age Glass in the Mediterranean,” Mediterranean Archaeology and Archaeometry, v. 10, no. 1, 2010, pp. 1–24; Kalliopi Nikita and Julian henderson, “Glass Analyses from Mycenaean Thebes and Elateia: Compositional Evidence for a Mycenaean Glass Industry,” Journal of Glass Studies, v. 48, 2006, pp. 71–120;

FIG. 6. Cross section of glass fragment shown in Fig-ure 5, showing turquoise color (Y 937-TRQ). FIG. 7. Scanning electron microphotograph (SE im-

age) showing stratigraphy of glass sample shown in Figure 5: (1) core of glass, (2) surface glass layer, and (3) external yellowish layer of ceramic remains.

FIG. 8. Backscattered electron microphotograph of area examined in Figure 7, showing distribution of antimony inclusions (white dots) in turquoise and green glasses.

concentrations of iron and aluminum oxides (~0.5% and ~1% respectively) are an indication of the use of either high-quality silicate sand or quartz pebbles as the network former.20 The low concentration of iron oxide and the negligible amount of cobalt suggest that the translucent turquoise glass (Y 937-TRQ) was colored with copper.

A comparison of the composition of the chunk of raw glass from Rhodes with that of other published translucent turquoise glasses of the same period21 revealed similarities with sam ples BRAK1 and BRAK16 from Tell Brak, syria.22 These samples have similar concentra-tions of sodium, magnesium, potassium, anti-mony, cal cium, iron, and copper oxides. Thus the glass from Rhodes shows characteristics re-flected in glasses imported from Mesopotamia.

In sample Y 937-Gr, the elevated level of alu-minum oxide (~8%) and the decreased quantity of silicon indicate the use of silica sand with

Julian henderson, “scientific Analysis of Glass and Glaze from Tell Brak and Its Archaeological Implications,” Excavations at Tell Brak, v. 1, The Mitanni and Old Babylonian Periods, ed. david Oates, Joan Oates, and helen Mcdonald, London: Brit-ish school of Archaeology in Iraq, and Cambridge: Mcdonald Institute for Archaeological Research, 1997, pp. 94–100.

22. henderson, Evans, and Nikita [note 21], pp. 8 and 11–13.

31

Sample Y 937-TRQ Y 937-TRQ (mineral

inclusion)

Y 937-Gr Y 937-Gr (mineral

inclusion)

Y 937-CerCeramic

Y 937-CerCeramic (mineral

inclusion)Oxide (wt %)

Na2O 14.3 10.2 12.1 10.8 4.8 0.8

MgO 3.7 2.8 3.1 2.8 5.4 2.1

Al2O3 1.1 0.7 7.7 6.4 13.1 4.4

siO2 60.8 42.8 57.7 51.5 48.1 27.9

P2O5 0.6 0.2 0.3 0.2 0.2 nd

sO3 0.7 0.3 0.6 0.5 1.9 0.2

Cl2O 1.1 0.7 0.6 0.6 1.2 0.1

K2O 2.9 1.8 3.5 3.0 2.2 0.4

snO2 0.1 0.3 0.3 0.5 0.1 nd

sb2O5 3.5 25.6 2.2 9.7 3.7 3.8

CaO 8.5 11.8 5.7 7.1 12.1 9.8

BaO 0.2 nd 0.3 0.2 0.3 nd

TiO2 0.1 0.2 0.4 0.6 0.5 6.3

V2O5 nd nd nd nd nd nd

Cr2O3 0.1 0.1 0.2 0.2 0.1 0.2

MnO 0.1 0.1 0.1 0.2 0.2 0.3

Fe2O3 0.5 0.3 3.5 2.6 6.3 39.3

CoO 0.06 0.03 0.02 0.02 nd nd

NiO 0.1 0.1 0.1 0.1 0.1 0.3

CuO 1.1 0.8 0.6 0.7 0.1 0.3

ZnO 0.1 0.1 0.1 0.3 0.1 0.3

As2O3 nd nd nd nd nd nd

PbO 0.1 0.8 0.4 1.1 0.4 2.0

Values around 0.1 wt % are considered semiquantitativend = not determined

TABLE 1

EdX Analysis of Fragment Y 937 from Chunk of Raw Glass(wt % of oxides)

32

several clay impurities. As in the Y 937-TRQ glass sample, the opacifier here must have been calcium antimonate, resulting in translucency and higher concentrations of antimony and cal-cium. By contrast, however, the green glass (Y 937-Gr) was colored by iron, while the copper oxide concentration (~ 0.6%) had little or no effect on the color of the glass.

On the outer surface of Y 937, a thin layer (about 1 mm) of ceramic material is preserved. This should be attributed to the ceramic crucible that was used to melt the glass.23 The examina-tion of the cross section under sEM reveals an interaction interface with the underlying glass layer. This indicates either an accident dur ing production or an unfinished surface of the glass chunk. The ceramic body is highly calcareous, exhibiting extensive vitrification. There is a con-siderable concentration of antimony (~3.5%), which was probably diffused from the batch and incorporated in the ceramic body dur ing melting and vitrification.

Conclusion

This analysis has suggested that the trans lu-cent turquoise fragment from a chunk of raw glass (Y 937-TRQ) colored by copper has a com-position conforming to that of Mesopotamian glasses. The proposed Mesopotamian origin of this ingot from Rhodes is very significant. It should be viewed within the broader context of intensive commercial activity that took place in the Aegean, and especially on Rhodes, during the 14th and 13th centuries B.C. This era was characterized by a thriving glassworking indus-try on Rhodes and by the presence of many ar-tifacts, both locally made and imported, in the Rhodian agora. These included Rhodo-Myce-naean (or Koan) and Cypriot pottery, which is very distinctive; bronzes and Cypro-syrian cyl-inder seals; and exotica from the Levant, such as ivory from northern syria, precious and semi-precious materials from the Orient, and raw glass of Mesopotamian origin.24

23. Thilo Rehren and Edgar B. Pusch, “New Kingdom Glass-Melting Crucibles from Qantir-Piramesses,” The Journal of Egyp tian Archaeology, v. 83, 1997, pp. 127–141.

24. Marketou [note 1], p. 787; Karantzali [note 4], pp. 149–150.