Weight systems and the emergence of the first Pan-European money

Nicola Ialongo et al, A small change revolution. Weight systems and the emergence of the first Pan-European money, Journal of Archaeological Science (2021). DOI: 10.1016/j.jas.2021.105379

Highlights

• Metal trade in Europe increases in the course of the Bronze Age.

• Systematic fragmentation of metal objects increases in the course of the Bronze Age.

• The spread of weighing technology is correlated to the spread of fragmentation.

• Cosine Quantogram Analysis and Monte Carlo simulations show that metal fragments comply with weight systems.

• Metal fragments were likely used as money.

Abstract: In the Bronze Age (c. 2300–800 BC), European communities gave up their economic independence and became entangled in a continental trade network. In this paper, we will test the hypothesis that the adoption of a ‘Pan-European’ currency has favoured the development of such a network. We define a methodology to test the money-hypothesis in pre-literate economies, based on analogies with the material characters of metallic money in the Ancient Near East. The statistical properties of metals from European hoards are compared with those of balance weights, in order to test the following expectation: if they were used as money, complete objects and fragments are expected to comply with standard weight systems. The results meet the expectation, and indicate that bronze fragments possess the same statistical properties as hack-silver money in the Ancient Near East. The sample includes approximately 3000 metal objects, collected from two test-areas: Italy and Central Europe. The sample of balance weights includes all the items known to date for pre-literate Bronze Age Europe, collected within the framework of the ERC Project ‘Weight and Value.’

Popular version: Scrap for cash: Bronze Age witnessed revolution in small change across Europe (phys.org)

4. Discussion

4.1. Weight regulation

The results of the statistical analysis support the money-hypothesis for Late BA Europe, showing that metal objects were probably intentionally fragmented in order to comply with weight systems. The analogy with the Ancient Near East suggests that bronze fragments in Europe had the same function as silver fragments in Mesopotamia, i.e. they performed the basic functions of money. The joint analysis of balance weights and bronze objects suggests that monetary patterns of exchange existed in BA Europe, that they complied with a Pan-European index of value, and that they were based on the use of metals as standard media of exchange. The actual relevance of the phenomenon is difficult to quantify. For the time being, we can only observe that metal fragments were used as money frequently enough to leave measurable traces in the archaeological record; how frequently is still not possible to define. Unlike proper balance weights – that produce sharp, neatly separated clusters of mass values (Ialongo and Rahmstorf, 2019) – bronze fragments produce small clusters that stand out from a diffused background noise. FDA suggest that weight-regulated fragmentation is not very accurate. Based on the statistical analysis, we can derive that systematic fragmentation tends to produce fragments with mass values that are multiples of weight units, and that complete objects and ingots were cast with no specific mass prescription. Just like in the Mesopotamian world, the compliance with weight systems is an indirect consequence of trade, rather than a pre-defined regulation. Assessing the actual relevance of the phenomenon will require further research and a larger number of regional samples.

As to why fragments tend to assume regular mass values, we can propose a hypothesis. The typical trade situation – documented in Mesopotamian texts – includes two agents, each provided with their own weighing equipment (Peyronel, 2011). Metal fragments do not need to be weight-regulated, since each agent can easily quantify the value of the transaction. However, pre-weighed fragments would speed up the operations, by preventing the calculation of a remainder. Breaking bronze objects is relatively simple, and does not require any particular metallurgical knowledge. Experimental results show that a socketed axe made of tin bronze (8%) breaks into pieces with three blows if heated up at c. 560°, the temperature of a medium-sized campfire (Knight, 2017). The required temperature is lower if copper is alloyed with lead (Knight, 2019). In order to produce accurate fragments, one can progressively break off small bits and repeat the weighing until the desired mass is obtained. The repetitiveness of the operation increases the skill of the operator, and skill increases accuracy. Fragmentation can be performed either before or during the transaction, depending on the situation. Archaeological evidence suggests that, at least in the Late BA, bringing along small stocks of metal fragments was a rather common habit. Many burials in central and northern Europe include small boxes of organic materials, containing metal fragments and scraps, blunt tools suited for breaking metals and, sometimes, scale beams and balance weights (Pare, 1999; Roscio et al., 2011). One of such boxes was recently identified among the finds of the Late BA battlefield in the Tollense Valley (Northern Germany, c. 1350–1200 BC) (Fig. 2), which attests that their use was not limited to the burial rite (Uhlig et al., 2019). These containers offer a suggestive picture of how metallic money could be carried around for everyday purposes.

For the complete objects, the absence of weight-regulation has a perfectly logical explanation. The subsample between 7 and 200 g includes every type of ornament, tool and weapon with the only exception of swords, which are exclusively represented in fragments. Since the size of a useable object is dictated by its function, there is no reason to assume that its mass should be regulated in order to fit a predetermined value.

CQA does not give significant results for ingots and ingot fragments in the shekel-range. This can depend on several factors. The fact that ingots are, on average, thicker than any other object in the sample could imply that they are more difficult to break into pieces with a predetermined mass. It could also mean that ingots and ingot fragments were not used as currency. Ingots are usually made of pure copper (e.g. Pernicka et al., 2016), and thus might have been used exclusively as raw material. It should be noted, however, that we could not compare the ingots to the balance weights in the mina-range. It is possible that, since they are on average significantly more massive than other object categories, their mass was measured in fractions of the mina rather than in multiples of the shekel. Finally, one has to consider that the main use of heavy balance weights is to weigh out bulks of goods, rather than single objects. The fact that individual ingot fragments do not comply with weight systems does not rule out the likely possibility that they were traded in weighed bulks. One way to test this could be to analyse the total weight of hoards, and verify if they conform to multiples of the European mina. Unfortunately, it is not easy to determine if a hoard is in pristine conditions, or if it underwent modifications before or after its recovery.

4.2. The value of money

Our results show that the value of bronze was quantified according to a shared frame of reference. What we define as ‘bronze,’ however, is an umbrella term for many different copper alloys, mostly containing variable proportions of tin and lead. It is theoretically possible that different alloys had different values, which could somehow hamper the circulation of metals as money. For BA Europe the puzzle is difficult to solve, as we have no direct evidence of value equivalences between different types of bronze and between bronze and other commodities.

An argument in favour of the hypothesis of the different values could be that, since bronze fragments are made of different alloys, their indiscriminate use as currency would hamper or entirely prevent their reuse as raw material. The argument, however, fails to correctly account for the evidence. Substantial quantities of fragments exist in the archaeological record and fragments were undoubtedly exchanged, regardless of whether or not one accepts their monetary use. At the same time, it would seem that metallurgists did not have problems in finding the right alloy for their purposes. This either implies that they were able to determine the alloy of fragments, or that they used fragments in limited quantities and mostly relied on other forms of raw metal, such as ingots. Hence, the argument also fails to acknowledge the relevant role of recycling (e.g. Radivojević et al., 2018). A second argument could be that some metals used in bronze alloys are rarer than others, and thus more expensive. Lead, for example, is alternatively assumed to be expensive (e.g. Johansen, 2016) or cheap (e.g. Needham and Cook, 1988), depending on whether it had to be imported (Scandinavia) or was locally available (British Isles). But if distance played such a determinant role, how can one explain, for instance, the rich metallic record of Denmark, which completely lacks copper and tin sources? The problem of value is extremely complex, and bears far-reaching implications; it basically implies theorising a continent-wide market economy for BA Europe, and cannot be addressed here. Hence, while these arguments are worthy of consideration, they should not play, for now, a decisive role in the interpretation, in one way or another.

An alternative solution could be to assume that the problem of value is not correlated to the diversity of alloys, but rather to different modes of circulation. Different models on the circulation of metals tend to focus perhaps too much on metallurgy: If fragments were mainly exchanged as money their main purpose was to circulate, not to be recycled. Moreover, their circulation was certainly not limited to metallurgists. In theory, a single fragment could circulate for decades without ever ending up in a metallurgist's hands.

Economic theory does not explain why money has value. One way to justify why worthless pieces of paper can have the same monetary function of precious metals (the so-called ‘Hahn's problem’; Hahn, 1965) is to admit that ‘if people believe that money has value, it does’ (Velde, 2021: 201). Once money is acknowledged to be valuable, however, the market dictates how much value money has. It follows that, if money can be a worthless substance, its market value is not necessarily correlated to the substance of which money is made. Since the value of money (and commodities alike) is regulated by the market, that value will be determined by the most frequent use that the market makes of the substance of which money is made. Hence, if metallurgy is the prevalent use, then it is possible that different alloys will have different values. If monetary use is prevalent, then the difference of alloys will play a minor role in the determination of value.

4.3. The origin of money? Towards a theoretical framework for money in BA Europe

Weighed currency was not the first form of money in Europe. It was, however, the first that could be potentially accepted anywhere, provided that weighing technology was available. In other words, weighing technology does not originate money, but simply provides a universally-valid frame of reference for the quantification of its value (Rahmstorf, 2016). The idea that the origin of money is correlated to technology or to the complexity of socio-economic systems implies an evolutionary paradigm, which hampers our understanding of the functions of pre-modern economies. There is nothing in economic theory that prevents the emergence of money in any given market, in economies of any complexity, and at any point in history (e.g. Jones, 1976; Velde, 2021). Economic evolutionism has been shown to be based on a substantial misunderstanding of the social dynamics that regulate the modern economy (Bloch and Parry, 1989). On the contrary, it has been contended that the modern western economy is still as much embedded in social institutions as only ‘primitive’ economies were previously believed to be (e.g. Appadurai, 1986), and that the economy in ‘primitive’ societies is substantially less embedded than the evolutionary paradigm would predict (e.g. Granovetter, 1985). The contemporary approach to prehistoric money owes much to a seminal article by G. Dalton (1965), in which the author traces a sharp distinction between ‘primitive’ and ‘modern’ money. Soon after, however, J. Melitz – an economist – demonstrated that Dalton failed to acknowledge that supposedly ‘primitive’ and ‘modern’ monies have, in fact, the same functions and limitations (Melitz, 1970). Since then, the functional equivalence between ‘primitive’ and ‘modern’ money is generally accepted by most economists, economic historians, and economic anthropologists (e.g. Jones, 1976; Velde, 2021; Zelizer, 2000).

Money is not an evolutionary milestone, but simply a solution to the practical problem of the ‘double coincidence of wants’ (Jevons, 1875:3), stating that no one can trade with anyone who does not need or does not want whatever it is that they have to offer in payment. For example, if a pig breeder wants wheat and has only pigs to offer, they cannot obtain wheat if the crop farmer does not need or want pigs. The problem can be solved by agreeing upon using a third medium that everyone will eventually come to accept, as it is widely documented in many so-called primitive economies (e.g. Einzig, 1966; Pryor, 1977). Money is not inevitable but it is convenient, as it has no requirements other than being customarily accepted by most agents in a given market. Metals represent only a limited part of all the pre-coinage monies documented either historically or ethnographically, which include perishable materials such as barley (e.g. BA Mesopotamia: Steinkeller, 2004), textiles (e.g. Classic Maya: Baron, 2018), bark-cloth (e.g. Early Colonial West Africa: Pallaver, 2015), and dried fish (e.g. Medieval Iceland: Mehler and Gardiner, 2021). Money is merely a convention, whose embodied physical media can have intrinsic value (such as silver coins) as well as none at all (i.e. banknotes) (Velde, 2021).

Concerning BA Europe, money is often believed to appear in some forms with the spread of mass metallurgy. The so-called Ösenringbarren (a type of ingot-like objects shaped as open rings common in Central Europe in the Early BA, c. 2150–1700 BC) are a common example. They show a noticeable regularity in shape, mass, and composition (Lenerz-de Wilde, 1995), and are regarded by some scholars as evidence of the earliest money in Europe (e.g. Pare, 2013; Kuijpers and Popa, 2021). Since they exist several centuries before the introduction of weighing technology in Central Europe, their approximate regularity can be explained by relatively standardised moulds, and by the intuitive ability of experienced users to determine the approximate mass-equivalence of two objects simply by holding them in both hands (Kuijpers and Popa, 2021). If Ösenringbarren were indeed money – which we have no reason to doubt – their function was no different from the later, weight-regulated metal currencies. The difference is in their circulation. Since weighing technology was not available (until proven otherwise) there was no way to assess their value objectively or, for example, to calculate fractions and multiples. One can think of Ösenringbarren as a form of ‘fiduciary money’, i.e. a standard medium of exchange whose value is conventionally agreed upon, and whose intrinsic value is not relevant for the quantification of their exchange value. As it was recently proposed, fiduciary currencies can predate the emergence of commodity currencies, contrary to the common belief (Bresson, 2021). Whether classifying different types of money may or may not be the point, we would like to draw attention on the reasons why some monies are accepted in some regions and not in others. The Ösenringbarren relied on their recognisable shape, approximate size, and peculiar chemical composition in order to be accepted, because these characters were well-known and understood in the limited area where they were in common use, i.e. between southern Germany and the Czech Republic. The reason why we do not find Ösenringbarren outside this area is because those same characters were not recognised as ‘valid’ elsewhere. The spread of metallurgy undoubtedly expands trade networks, and increases the potential utility of money. The introduction of weighing technology, on the other hand, exponentially expands the user base by providing an objective frame of reference that transcends traditional cultural boundaries. At the same time, weighing technology does not alter the functions of money, and does not imply the erasure of other patterns of monetary and non-monetary exchange. Finally, if weighing technology is not a requisite for money, neither is metallurgy. By the same token, we should not prejudicially rule out a wide range of perishable commodities that are invisible in the archaeological record, but which could have been used as money before, during and after the introduction of metallurgy.