by Felix Martin
A state with such a complex, hierarchical, and bureaucratic form of organisation demanded altogether different technologies of social co-operation and control from the primitive institutions that governed the small, tribal societies of Dark Age Greece. It is therefore not surprising that Mesopotamia witnessed the invention of three of the most important social technologies in the history of human civilisation: literacy, numeracy, and accounting.
THE SILICON VALLEY OF THE ANCIENT WORLD
The ancient world was mystified by the origins of literacy. It seemed inconceivable that a technology so self-evidently fundamental to civilised life could have been dreamt up by feeble-minded mortals. The only possible explanation, therefore, was that it had come from the gods—either as a generous gift, or as stolen goods. The Egyptians, for example, believed that Thoth, the baboon-faced god of knowledge, presented writing to mortals, and the Greeks that Prometheus did the same. In ancient Mesopotamia, on the other hand, it was held that the secret of literacy had been acquired by stealth. The great goddess Inanna had stolen writing for her city of Uruk by taking advantage of Enki, the god of wisdom, while he was drunk.
When modern scholars began to show an interest in the same question in the eighteenth century, they demonstrated more confidence in the powers of human invention. Archaeological evidence was marshalled, and by the early twentieth century a reasonable theory, consisting of two hypotheses, had been constructed. First, writing had not evolved gradually, but had been invented—exactly by whom was unclear, but it was generally presumed to be by wise sages who “agreed upon a conventional method of recording [language] in written signs … intelligible to all their colleagues and successors.”16 Second, the earliest writing must have been “pictographic”—that is, it consisted of stylised pictures of what it was intended to represent—since it would otherwise have been difficult for the sages to agree on the symbols and disseminate them easily amongst the population.17
Until the early twentieth century, all available evidence seemed to corroborate this pictographic theory of the origins of literacy. The earliest writing indeed appeared suddenly in the archaeological record, and did so in the form of Egyptian hieroglyphs, ancient Chinese characters, and the colourful pictograms of the pre-Columbian Aztec codices. In 1929, however, a new discovery turned the theory on its head. Excavations at Mesopotamian Uruk uncovered a vast archive of clay tablets inscribed with detailed accounts of palace and temple transactions. Dating from the late fourth millennium bc, the writing on these tablets represented by far the oldest specimens ever discovered. But unlike the pictographic scripts of Egypt, China, and Central America, this script consisted of abstract signs composed of combinations of inscribed flicks of a reed pen—a so-called “cuneiform” script. As the excavations continued throughout the mid-twentieth century, more and more evidence accumulated. The earliest known writing was represented not by pictograms but by a script not qualitatively different from a modern alphabet. If the conventional theory’s second hypothesis was incorrect, perhaps its first hypothesis of spontaneous invention was also flawed. All of a sudden, the origins of writing were plunged back into obscurity; four decades would pass before a new light was to illuminate the ancient question from a quite unexpected angle.
The interwar excavations at Uruk that discovered the earliest known writing were part of a golden age of archaeological exploration in Mesopotamia from the late nineteenth to the mid-twentieth century. American, German, and British campaigns throughout this period unearthed the sites of countless archaic settlements and yielded vast and impressive archives of the craftsmanship of the Mesopotamian civilisations, from monumental statuary to delicate jewellery. In amongst these highly prized finds, however, were also scattered many thousands of small clay artefacts—most of them no larger than children’s marbles. They came in many shapes and sizes—cones, cylinders, balls—but were otherwise utterly nondescript. For decades, these uninspiring-looking bits of debris were therefore largely ignored by archaeologists. Until the 1970s there was not even general agreement on what they were. Typical speculations were that they were “children’s playthings,” “amulets,” or “game pieces.”18 Often they were identified simply as “objects of uncertain purpose.” One distinguished American archaeologist wrote in his site report that “[f]rom levels 11 and 12 come five mysterious … clay objects, looking like nothing in the world but suppositories.”19
The truth was both more mundane and more momentous. In 1969, a young French archaeologist, Denise Schmandt-Besserat, decided to make a comprehensive catalogue of these mysterious bits of clay. Once analysed together, it became clear that they came in various generic shapes and sizes common to sites from all over Western Asia, from south-eastern Turkey to present-day Pakistan. Schmandt-Besserat realised that these long-overlooked artefacts were not primordial chess pieces or primitive laxatives but tokens used for what is called “correspondence-counting”—keeping track of one quantity by maintaining a matching quantity of something else. Correspondence-counting requires no numerical sophistication whatsoever, merely the ability to check whether two quantities are the same.20 It is the earliest known technique for reckoning number: notched animal bones thought to use correspondence-counting to record the passage of days or the number of animals killed have been discovered dating from the early Stone Age.21 In Mesopotamia, Schmandt-Besserat realised, a complex system of clay tokens had enabled this ancient method to attain a previously unknown level of sophistication. Each different shape and size of token represented a different type and quantity of a particular staple commodity: incised cones for bread, ovoids for oil, rhomboids for beer, and so on.22 Reckoning number using this elaborate system of tokens was put to use in the agricultural economy to keep account of numbers of animals or quantities of crops.
For thousands of years, this system remained unchanged in its essentials.23 With the rise of urban civilisation and the temple economies, the demands on record-keeping increased dramatically. Around 3100 bc, in Mesopotamian Uruk, a critical innovation was made. Records began to be kept not using collections of the tokens themselves, but by making impressions of the tokens on moist clay tablets. Henceforth, a sheep would not be represented by a conical token kept in an account box, but by the triangular impression of such a token on a clay tablet. Once this system had been introduced, and the impressions corresponding to each token had been learned, it was a small step to dispense with the tokens themselves. An impression of the correct shape and size could be made in the wet clay of a tablet much more simply using a reed pen. The ancient system of three-dimensional objects had been translated into a new system of two-dimensional symbols. It was an epochal development: nothing less than the birth of literacy.
Stimulating the invention of writing was no mean achievement on its own; but the increasing complexity of the Mesopotamian economies meant that the pressure to devise ever more efficient and flexible techniques was unrelenting. Reckoning number using the new, written symbols was certainly more efficient than shaping, firing, and then storing thousand upon thousand of little clay tokens. But both techniques still relied upon correspondence-counting—one token or symbol corresponding to each thing being counted. Soon after the invention of writing, however, another momentous improvement was made. Instead of writing five sheep symbols to signify five sheep, separate symbols for the number five and the category sheep were introduced. Now, only two symbols were required, instead of five. When one considers that on a single surviving tablet the receipt of 140,000 litres of grain is recorded it is obvious that the practical advantages were considerable.24 The longer-term implications were even greater, however. Correspondence-counting requires no notion of abstract number; no concept, that is, of number separate from the things being counted. The new system did. Not only had Ur invented writing, it had almost simultaneously invented the concept of number—and thereby opened the way to the development of mathematics.
The invention of writing and abstract number set the stage for the developmen
t of the third technology at the heart of Mesopotamian society: accounting. The hierarchical control of economic activity by clerical bureaucracies required a management information system: a technique for quantifying stocks and flows of raw materials and finished goods, for using these quantities in forward planning, and for checking that the plan was being correctly carried out on the ground. Accounting was a social technology that combined the ability to keep records efficiently using writing and number with standardised measures of time so that quantities could be tracked as stocks on balance sheets and flows on income statements.25 For the economies of ancient Mesopotamia, as for large corporations today, it was a system of consistent book-keeping that allowed directives from on high to be translated into practical instructions—and for the fulfilment of those instructions then to be verified by that most familiar, most forbidding, and, as it transpires, most ancient of professional figures: the accountant.
Thus in almost every respect, the societies of ancient Mesopotamia represent a radical counterpoint to those of Dark Age Greece. In place of the primitive and egalitarian tribal society of Homer there was the city, with tens of thousands of inhabitants ruled by a semi-divine king and organised into a multi-layered hierarchy. Instead of the exercise of raw power by chieftains over commoners, there were the sophisticated rules of the accounting system administered by the temple bureaucracy. In place of a simple economy governed by principles of reciprocity and ritual sacrifice that would have been familiar to countless primitive tribes over the past several millennia, there was a complex economy governed according to an elaborate system of economic planning that would be familiar to a manager in a modern multinational corporation. Yet despite these yawning differences, there was one vital respect in which the economies of ancient Mesopotamia were identical to those of Dark Age Greece. For neither the bureaucratic plan of the temple, nor the primitive tribal institutions of Dark Age Greece, had any use for money.26
Why was it that this extraordinary commercial civilisation, the most advanced economy that the world had ever seen, the society that invented literacy, numeracy, and accounting, did not invent money? The answer is that it did not develop one critical ingredient—the single most important precondition for money and its central component. To understand what that ingredient was, we must take a detour to a bureaucratic environment of a much more recent vintage: the 11th meeting of the General Conference on Weights and Measures, on 14 October 1960, in Paris.
GETTING THE MEASURE OF THINGS
Faceless international bureaucracies have not typically been responsible for revolutionary advances in human civilisation. More often, they have been bastions of dogma and recidivism against which lonely pioneers have had to struggle in the daring quest for knowledge and truth. The field of metrology—the science of measurement—provides a notable exception to this general rule, however. On 14 October 1960, the quadrennial General Conference on Weights and Measures was convened to consider a set of proposals made by the International Committee for Weights and Measures, received by them from the International Bureau of Weights and Measures. It was as impressive an accumulation of faceless international bureaucracies as one could wish for—a certain recipe, one might have thought, for a turgid agenda of incidental points of order to be pored over in tedious detail by the delegates before adjournment for a long lunch on expenses. Nothing could have been further from the truth. For at this meeting was agreed, for the first time in history, a simple and universal system of units of measurement based on internationally agreed standards—the Système International d’Unités, or SI for short.
This was no small feat. Until the nineteenth century, consistent standardisation of units of measurement across any wide geographical area was virtually unheard of. In 1790, for example, a survey was commissioned to ascertain the standard length of the arpent, a common French unit of length. To the surveyors’ dismay, they found nine different standards in use in the département of Basses-Pyrénées alone. In Calvados, there were no fewer than sixteen.27 Nor were examples such as these by any means the most extreme: France was at the more enlightened end of the European spectrum when it came to consistency. “Altogether, a state of shocking confusion reigned,” wrote the great metrological scholar Witold Kula of his native Poland: “in the single village of Jastrzebie, Upper Jastrzebie used the Pszczyna measure while Lower Jastrzebie used the measure of Wodzislaw, and the vicar kept both measures available until the 1830s.”28
Then there was the proliferation of units themselves. Under the SI, length—any length, of anything—is measured by the metre, or its subdivisions or multiples. Metrological concepts applicable in such universal contexts were unknown in medieval and early modern Europe. Even today in the U.K., whisky is measured in gills, beer in pints, and petrol in gallons. But in the system of old Slavonic measures, for example, the foot was the length unit employed to measure out potato patches, while the pace was used to describe distances to be travelled. The fathom was used to record the depth of the sea, while the ell was used in the measuring out of cloth. Of course, what was actually being measured in all these cases was length. But a different unit was used for each specific context. This hodgepodge of vernacular units resulted in terminology which sounds almost nonsensical to modern ears: “[t]he peasant fisherman would refer to his net as being 30 fathoms long and ten ells wide.”29
This was the lamentable state of affairs which the General Conference on Weights and Measures had been established to remedy, and the creation of the SI was the culmination of nearly a century’s worth of international efforts to simplify and standardise the world’s weights and measures. It was a revolutionary advance in both respects. With regard to simplification, it introduced a set of just six basic units, sufficient for the measurement of any aspect of the physical world: the metre for linear extension, the kilogram for mass, the second for time, the degree Kelvin for temperature, the candela for luminosity, and the ampere for electric current.30 Its achievements in standardisation were even more dramatic. Not only did it establish internationally agreed standards for these basic units, but for the first time it defined them in terms of universal constants found in nature, rather than by reference to particular agreed examples. Henceforth, the SI metre, for example, was no longer defined in terms of a canonical metre rule kept in Paris, but in terms of the wavelength of radiation emitted by a particular chemical element.31
At first glance, this long march towards simplification and standardisation might seem to have been purely cosmetic. After all, regardless of their specific origins, even archaic units of measurement are all related to one another, and to modern units, in fixed proportions. What could be more harmless than the indulgence of local custom—or more typical of a faceless international bureaucracy than the urge to eradicate it? But this would be to misunderstand the nature and origins of systems of measurement. After all, the question can be reversed. Why did anyone ever settle for these limited-purpose units of measurement, when they could have had universally applicable ones? Why, in other words, did these absurd proliferations of local and limited-purpose units spring up in the first place?
The truth is that there was method in the apparent madness. The common feature of traditional metrological concepts was that they had been developed from the bottom up for use in specific contexts—and that they captured exactly the most relevant aspect of the activity at hand. Today, for example, we define the area of any piece of land by measuring its perimeter. To the medieval peasant farmer, however, the square dimensions of a piece of arable land were the least useful things to know about it. Instead, as Witold Kula explained, “[t]wo qualitative aspects of any cultivable field are of crucial importance: the time it takes to cultivate it, and the harvest it is capable of yielding.”32 As a result, traditional units for the measurement of agrarian land were typically defined in terms of the area that one man could plough in one day, or that would yield a given volume of grain. The square dimensions of units so defined may of course vary significantly wi
th the quality of the land; but what seems to the modern mind an unfortunate loss of generality was at the time a gain in terms of precise usefulness for the task in hand. The example illustrates a general point: the appropriate extent and standardisation of any metrological concept depends upon its use.
An early attempt at simplification and standardisation: a French cartoon of 1795 explaining the merits of the new “metric” system.
(illustration credit 2.1)
Of course, metrology is not static: as the uses to which they are put evolve, so do units of measurement and their standards. What is more, there is feedback in both directions: if practices and techniques give rise to units of measurement, the invention of broader metrological concepts and the implementation of more consistent standards allow new forms of technological and economic co-operation to flourish. Myriad inconsistent systems of measurement and standards that varied village by village might have been sufficient for an economy of isolated agricultural smallholdings; but the industrial age—the age of machines and of mass production—demanded standardisation, and the burgeoning of international trade and industry demanded common units in the name of efficiency. Today, the need for universal units calibrated to common standards is more acute than ever. In August 2011 The Economist magazine analysed the origin of the 178 components that make up the Apple iPhone 4: a quarter were from South Korea, a fifth from Taiwan, a tenth from the United States, and other fractions from Japan, China, and a host of European countries.33 Global industrial supply chains—to say nothing of international collaboration in medicine, science, and commerce—would be inconceivable without globally understood units of measurement. None of it could exist if the process crowned by the creation of the SI had not taken place.