NATURE, VOL 411, 24 MAY 2001, p.437

www.nature.com

Converting currencies in the Old World

Simple arithmetic underpinned trading throughout the Near East during the Bronze Age.

Raw materials recovered from archaeological excavations in the Indus Valley, the Persian Gulf, Mesopotamia, Egypt and the eastern Mediterranean reflect the existence of long-distance trading during the Bronze Age, which united these regions into networks of commercial exchange. As each region relied on a different set of weights for trading, a straightforward conversion system must have been in operation. Here we describe a simple and universal conversion system that could have provided an economic key to the trade networks of the Old World between 2500 and 1000 BC.

Old World trading networks brought raw resources such as lapis lazuli, copper, tin, silver and gold, as well as exotic materials and artefacts such as cylinder seals, decorative pottery and carved stone vessels from far afield. Establishing equivalence between weight systems that were peculiar to different regions would have created comparable units of value as well as mutual confidence in buying and selling goods, which would have been essential for dealing with universally precious materials such as gold and silver.

Many different weight systems were in operation in the Bronze Age (see Fig. 1 for an example), all of which were more or less amenable to a 'standard' system of conversion. A Mesopotamian text [1] from about 1800 BC mentions the conversion of a weight from the Dilmun standard (used on the islands and northeastern shores of the Arabian peninsula) to the Old Babylonian Ur standard: "... they had given to us 5,532+2/3 minas of copper according to the standard of Dilmun. Its weight is in total 611 talents 6+2/3 minas according to the standard ofUr."

Different approaches have been used to study ancient weight systems [2-9] � for example, attempts have been made to trace a linear development into our prevailing metrological systems [2]. We base our investigation on the use of talents, shekels and minas as units of weight, although their actual weights vary in different systems (see Table 1 of supplementary information). For example, a Syrian talent is 28.2 kg, which is subdivided into 60 minas of 470 g each; a mina is subdivided into 50 shekels of 9.4 g. Thus, in the Ugarit system, 3,000 shekels of 9.4 g (3,000 X 9.4 g = 28,200 g) would be the same as a Syrian talent of 28.2 kg.

This Ugaritic shekel of 9.4 g can be interrelated with all other weight systems in the Near East and the Mediterranean. It is identical to the Egyptian kdt of 9.4 g and is equal to three-quarters of the Egyptian gold dbn of 12.83 g. Futhermore, four Ugaritic shekels (4 X 9.4 g = 37.6 g) is roughly equivalent to three Egyptian gold dbn (3 X 12.83g = 38.49g). We doubt that the weight systems in the second millennium BC were accurate enough to detect an error of less than 1 g in 38 g, or 2.6%.

Figure 1. Stone weights in the shape of ducks from the archaeological site of Yorghan Tepe, ancient Nuzi, Iraq. Nuzi was a provincial town of the Human Kingdom (1600-1400 ��). The weights represent 20 shekels (smallest), 1 mina and 2 minas (largest), and are marked with Incisions to denote their value. (Photo: Semitic Museum, Harvard University.)

These equivalences allowed the use of weights in different combinations, generally multiples of two or five. For example, two weights each with the value of two Ugaritic shekels were equivalent to two weights of the value of one and two Egyptian dbn. Thus, two Ugaritic shekels (9.4g X 2 = 18.8g) plus two more Ugaritic shekels comes to 37.6 g, and two weights with the value of one (12.8 g) and two Egyptian gold dbn (25.6 g) comes to 38.4g. In another example, 9.4g of the Ugaritic shekel equals four-fifths of the Hittite shekel of 11.75g (5 X 9.4g = 47 g and 4 X 11.75g = 47g), so five Ugaritic shekels equals two weights each with the value of two Hittite shekels.

This conversion system enables a hypothetical 1,370-g 'ingot' of lapis lazuli to be defined in terms of the different shekel weights of each region. Thus, 1,370 g would be the equivalent of 100 13.68-g Dilmun shekels , 160 8.55-g Mesopotamian shekels, 175 7.83-g Eblaite-Carcemish shekels, and so on, across an east-west route to the Mediterranean (see Table 2 of supplementary information, which also uses a tin ingot as an example).

Our analysis of the diverse weight systems that were in use from the Indus Valley to the Aegean from the middle of the third millennium to the end of the second millennium �� complements earlier studies [6-8] and reveals that the conversion systems in operation were elegant in their simplicity, although each weight system showed a degree of variance from our proposed standards. The integration of different weight systems was crucial in developing the scale and nature of commercial exchange in the Near East and would have facilitated the emergence of the Ancient World System [10].

Alfredo Mederos*+, �. �. Lamberg-Karlovskyt *Departamento de Prehistoria, Universidad Complutense, Ciudad Universitaria, 28040 Madrid, Spain

^Department of Anthropology, Peabody Museum, Harvard University, 11 Divinity Avenue, Cambridge, Massachusetts 02138, USA e-mail: karlovsk@>fas. harvard, edu

1. Roaf,M. frai)44, 137-141 (1982).

2. McDonald, D. M. The Origins of Metrology (McDonald Institute for Archaeological Research, Cambridge, 1992).

3. Bibby, �. G. KumI 1970, 345-353 (1970).

4. Powell, M. A. in Studies in Honour ofT. B. Jones(eds Powell, M. A. & Sack, R. H.) 71-201 (Kevelaer, 1979).

5. Edzard, D. 0. (ed.) Reallexikon der Assyriologie und Vorderasiatischen ArchaologieVo\. 7,457-517 (de Gruyter, Berlin, 1990).

6. Psnse.H.F.nialoghidiArchealogia 3,155-159 (1981).

7. Zaccagnini, �. in Trafftd Micenei' nel Mediterraneo (eds Marazzi, M. etal.) 303-314 (Istituto per la Storia e 1'Archeologia della Magna Grecia, Taranto, 1986).

8. Zaccagnini, �. II Congresso Internazionale di Studi Fenici e Punic. 343-347 (CNR, Roma, 1991).

9. Courtois, I. �. Res. Orientates 2, 119-127 (1990).

10. Frank, A. G. Curr. Anthropol. 34, 383^129 (1993).

Supplementary information is available on Nature's website at http://www.nature.com or as paper copy from the London editorial office of Nature.