The basic components of early glasses were soda, lime, and silica. In the Mediterranean region and Western Asia, these three components were usually introduced in the form of two ingredients: either natron (soda) and sand containing shell fragments (silica and lime), or ash derived from halophytic plants (soda and lime) and calcium-free sand or crushed quartzite (silica). The raw materials used by glassmakers in the Levant (the eastern littoral of the Mediterranean Sea) in the mid-first millennium A.D. were natron and calcium-rich sand collected on local beaches, the most famous of which was the beach at the mouth of the Belus (Na'aman) River, Israel. The raw materials used by most glassmakers in the same region half a millennium later were plant ash and calcium-free sand or crushed quartzite.¹ This note draws attention to events that may have influenced the transition from natron to plant ash. Although my focus is the Levant, the note may have wider implications.²

The most extensive deposits of natron in the Mediterranean region are in the Wādi el-Natrūn, between Cairo and Alexandria, in Egypt. The Wādi el-Natrūn contains a series of lakes that are rich in salt and natron. Lakeside deposits of these minerals have been exploited since antiquity,³ and it is generally accepted that the Wādi el-Natrūn was the source of the natron used by Roman and Byzantine glassmakers in Egypt and the Levant. Presumably it was exported by sea from Alexandria.

Natron in the Levant: The Latest Evidence

With one exception, the latest evidence for the use of natron by Levantine glassmakers consists of raw glass from the sixth- or seventh-century glass factory at Bet Eli'ezer (Hadera) and eighth- to ninth-century glass vessels from Ramla, both in Israel.⁴ Similarly, glass vessels from Jordan and southern Syria were made with natron at least until the early seventh century (the date of the latest samples).⁵ The transition to plant ash in the Levant, therefore, seems to have taken place in or after the eighth century. The change, it should be noted, was not necessarily universal; glassworkers at Sepphoris, Israel, are reported to have used natron continuously throughout the Roman, Byzantine, and Islamic occupation of the site.⁶

The switch from natron to plant ash was sometimes expensive. At Bet She'arim, Israel, glassmakers introduced too much lime in the batch, apparently by combining one traditional ingredient (calcium-rich beach sand) and one perhaps unfamiliar ingredient (calcium-rich plant ash), thereby causing the melt to fail. This failure resulted in the loss of nearly nine tons of glass.⁷ The magnitude of the loss suggests that the new ingredient was forced upon Levantine glassmakers by circumstances beyond their control.

Those circumstances may have been events in Egypt, which appear to have escaped the attention of students of glass history.⁸ The ruler of the Abbasid caliphate, Hārūn al-Rashīd, died in 809. Hārūn designated his eldest son, al-Amin, as his successor, and another son, al-Ma'mun, to succeed al-Amin. After he ascended the throne, however, al-Amin made clear his intention to nominate his own son to succeed him. ln Egypt, hostilities broke out between supporters of al-Amin and al-Ma'mun in 811. Although al-Amin died in 813 and al-Ma'mun became caliph, a more or less continuous state of civil war existed in Egypt until 832.

One aspect of the situation in Egypt deserves our special attention. In May 814, rioting broke out in a suburb of Cordoba in al-Andalus (Muslim Spain).⁹ After it had been suppressed, the emir of al-Andalus ordered the suburb to be destroyed and the inhabitants to be deported. Some embarked for western Africa, and others for Egypt. The latter, not counting women and children, are said to have numbered 15,000. They reached Alexandria, but the authorities prevented them from landing, and they lived on their ships, which were anchored offshore. In 815, however, a deposed governor of Alexandria enlisted their help in a revolt. The Andalusī seized this opportunity, looted the city, and killed the deposed governor. This last action led to a pitched battle outside Alexandria with an Arab army, which the Andalusi defeated. Meanwhile, the Alexandrians killed all of the Andalusī who had remained in the city. When the victorious Andalusī returned, they massacred the Alexandrians and put whole districts of the city to the torch. The Andalusī remained at Alexandria, maintaining their virtual autonomy, until 826, when they capitulated to an army led by al-Ma'mun's commander in Egypt, Abd Allāh ibn Tāhir. Under the terms of their surrender, the Andalusī left Egypt; subsequently, they created a new base, in Crete.

Thus, for 21 years, Egypt was in a state of civil war, and for 11 of those years, Alexandria, the principal port of Egypt, was occupied by Andalusī rebels. lt is not impossible that, under these circumstances, glassmakers outside Egypt were unable to acquire natron from time to time. In order to stay in business, therefore, Levantine glassmakers may have been compelled to adopt the raw materials used by glassrnakers farther east: plant ash and calcium- free sand or crushed quartzite.

It is sometimes said that chemical analyses of Egyptian coin weights—in particular the analyses conducted by Matson, Sayre, and Gratuze and Barrandon—show that glassmakers in Egypt also began to use plant ash in the first half of the ninth century.¹⁰ This statement, however, is mistaken. Matson reported just one analysis of a coin weight, which was struck in the 120s/737– 747 (perhaps in 123/740– 741); it was made with natron. Sayre reported the analysis of 13 coin weights, and although he did not provide either the exact dates of his samples or details of every analysis, he noted that five eighth-century weights and one ninth-century specimen had low-magnesium compositions consistent with being made with natron, while seven 10th-century weights had relatively large amounts of magnesium and so were made with plant ash. Gratuze and Barrandon analyzed 70 coin weights and vessel stamps, and they identified four compositional groups of glass, only one of which was made with plant ash. The earliest coin weight in the plant-ash group was made in 358/968–969. The latest coin weight made with natron was issued between 254/ 868 and 270/883–884. The three reports do not mention a single analysis of a coin weight identified as made between 868 and 968–969. Thus, according to the available evidence, natron was still being used in Egypt in 868 and its replacement or partial replacement by plant ash may have taken place between 868 and 968–969. Indeed, as far as I am aware, we have no evidence at present for the use of plant ash by glassmakers in Egypt before 968–969. As one would expect, the proposed disruption of Alexandria's overseas trade sometime between 811 and 832 had no immediate effect on the availability of natron in Egypt itself.

The Introduction of Plant Ash

Analyses of samples from an early islamic glass factory at Raqqa, Syria, revealed four different compositions, three of which contained plant ash, while the fourth was made with natron.¹¹ Coins indicated that the factory operated in the first half of the ninth century, and the excavator believes that it was already in use in 804 and ceased to operate in the first quarter of the ninth century. He notes that the finds are the earliest documented use of plant ash by Islamic glassmakers, and that they date "from a time when the plant ash technology was being introduced into the Islamic world."¹²

Regardless of whether plant ash was first used by glassmakers at Raqqa before or after 804,¹³ it is unlikely that this will tum out to be the first occurrence of plant ash for glassmaking in the entire Islamic world. Beginning in the Bronze Age, plant ash was employed continuously by glassmakers in the inland regions of the Middle East, where natron does not appear to have been used,¹⁴ and Sasanian and early Islamic glasses from Iraq and Iran consistently have the relatively high levels of potassium and magnesium characteristic of glass made with plant ash.¹⁵ Indeed, perhaps the most likely explanation of the appearance of plant-ash glass in the factory at Raqqa is the presence of glassmakers from some part of the Middle East, where plant ash was habitually used as the source of soda. In other words, it is not necessary to suppose that the glassmakers of Raqqa discovered for themselves that plant ash is a viable substitute for natron. Perhaps they simply began to make glass in the manner in which their neighbors down the Euphrates had been making it for the best part of 3,000 years.

Conclusions

l. From time to time between 811 and 832, civil war in Egypt may have made it difficult for glassmakers in the Levant to obtain natron. The situation may have been particularly difficult between 816 and 826, when the port of Alexandria was occupied by refugees from Cordoba.

2. During this period, a shortage of natron may have compelled Levantine glassmakers to use plant ash, and subsequently it became a standard ingredient. Glassmakers in the Levant may have learned about plant ash from glassmakers at Raqqa or some other place in Syria where plant ash (the traditional source of soda in Mesopotamia) was already in use.

3. Later (apparently between 868 and 968–969), glassmakers in Egypt also began to use plant ash.

David Whitehouse
This article was published in the Journal of Glass Studies, Vol. 44 (2002), 193–196.

Notes

Acknowledgments. I am indebted to Prof. Michael McCormick for drawing my attention to the events at Alexandria in 815-826, and to Dr. Jan C. Freestone for showing me the text of his forthcoming paper on the composition of early Islamic glass from Ramla, Israel.

¹ Robert H. Brill, "Scientific Investigations of the Jalame Glass and Related Finds," in Excavations at Jalame, Site of a Glass Factory in Late Roman Palestine, ed. Gladys Davidson Weinberg, Columbia, Missouri: University of Missouri Press, 1988, pp. 257–294, esp. pp. 264–269; Jan C. Freestone, Yael Gorin-Rosen, and Michael J. Hughes, " Primary Glass from Israel and the Production of Glass in Late Antiquity and the Early Islamic Period," in La Route du verre: Ateliers prima ires et secondaires du second millénaire av. J-C. au Moyen Age, ed. Marie-Dominique Nenna, Travaux de Ia Maison de l' Orient Mediterraneen, no. 33, Lyons: Maison de l' Orient Méditerranéen- Jean Pouilloux, 2000, pp. 65–83.

² At roughly the same time, glassmakers in western and central Europe also abandoned natron, replacing it with wood ash. See Karl-Hans Wedepohl , "The Change in Composition of Medieval Glass Types Occurring in Excavated Fragments from Germany," Annates de l’ Association lnternationale pour I'Histoire du Verre, v. 14, Venice/Milan, 1998 (Lochem, 2000), pp. 253–257; idem, "The Composition of Glass from the Carolingian and Post-Carolingian Period in Central Europe," in La ventata in Occidente dal IV all 'Xl secolo, ed. Francesca Dell ' Acqua and Romano Silva, Il colore nel Medioevo: Arte, simbolo, tecnica, no. 3, Lucca: Istituto Storico Lucchese, Scuola Normale Superiore di Pisa, and Corpus Vitrearum Medii Aevi ltalia, 2001 , pp. 257–270.

³ Strabo 17.1.22- 23; Pliny NH 31.106–122.

⁴ Freestone, Gorin-Rosen, and Hughes [note 1], pp. 71 - 72; Jan C. Freestone, "The Composition of Early Islamic Glass from Ramla, Israel," Atiqot, forthcoming.

⁵ Odile Dussart, Le Verre en Jordanie et en Syrie du Sud, Bibliothèque archéologique et historique, v. 152, Beirut: lnstitut Français d' Archéologie du Proche-Orient, 1998, p. 190.

⁶ Alysia Fischer and W. Patrick McCray, "Glass Production Activities as Practised at Sepphoris, Israel (37 BC–AD 1516)," Journal of Archaeological Science, v. 26, 1999, pp. 893–905, esp. pp. 901–903.

⁷ Ian C. Freestone and Yael Gorin-Rosen, "The Great Glass Slab at Bet She' arim, Israel: An Early Islamic Glassmaking Experiment," Journal of Glass Studies, v. 41, 1999, pp. 105–116. Similar failures at Arsuf in Israel and Tyre in Lebanon, which may or may not have been due to the same problem, resulted in the loss of more than two tons and at least 30 tons, respectively.

⁸ Gaston Wiet, L 'Egypte arabe de la conquéte arabe à la conquéte ottomane, 642- 1517 de l 'ere chrétienne, Histoire de la nation égyptienne, v. 4, ed. Gabriel Hanotaux, Paris: Société de l' Histoire ationale and Librairie Pion, 1937, pp. 64–74; Hugh Kennedy, "Egypt as a Province of the Islamic Caliphate, 641-868," in The Cambridge History of Egypt, v. I, Islamic Egypt, 640–1517, ed. C. F. Petry, Cambridge: Cambridge University Press, 1998, pp. 62- 85, esp. pp. 80–83.

⁹ Ibid. , pp. 68, 69, 71, and 72.

¹⁰ Julian Henderson, "Archaeological and Scientific Evidence for the Production of Early Islamic Glass in al-Raqqa, Syria," Levant, v. 31, 1999, pp. 225–240, esp. p. 234. The literature cited in support of this statement is Frederick R. Matson, "The Manufacture of Eighth-Century Egyptian Glass Weights and Stamps," in Early Arabic Glass Weights and Stamps, ed. George C. Miles, Numismatic Notes and Monographs, no. Ill , New York: The American Numismatic Society, 1948, pp. 31-69; E. V. Sayre, "Summary of the Brookhaven Program of Analysis of Ancient Glass," in Application of Science in Examinations of Works of Art, Proceedings of the seminar held at the Boston Museum of Fine Arts, Boston, Massachusetts, September 7- 16, 1965, Boston: the Museum of Fine Arts, 1965, pp. 145–154; and B. Gratuze and J.- N. Barrandon, " Islamic Glass Weights and Stamps: Analysis Using Nuclear Techniques," Archaeometry, v. 32, no. 2, 1990, pp. 155–162 .

¹¹ Julian Henderson, "Glass Production in Raqqa, Syria: New Evidence for an Interim Assessment," Annales de l 'Association lnternationale pour l 'Histoire du Verre, v. 13, Low Countries, 1995 (Lochem, 1996), pp. 257–268; idem, " Archaeological and Scientific Evidence for the Production of Early Islamic Glass in al-Raqqa, Syria," Levant, v. 31 , 1999, pp. 225–240.

¹² Henderson, "Glass Production" [note II], p. 267.

¹³ The difficulty of dating archeological deposits on the basis of coins that may have been old at the time of burial is a recurrent problem.

¹⁴ Robert H. Brill, "The Chemical Interpretation of the Texts," in Axel von Saldern and others, Glass and Glassmaking in Ancient Mesopotamia: An Edition of the Cuneiform Texts Which Contain Instructions for Glassmakers, with a Catalogue of Surviving Objects, Corning: The Corning Museum of Glass, 1970, repr. 1988, pp. 105–128.

¹⁵ Robert H. Brill, Chemical Analyses of Early Glasses, Corning: The Corning Museum of Glass, 1999, v. I, pp. 82–101 , and v. 2, pp. 151–205.

Published on February 27, 2013