People, Places, Events and Terms To Know:

I. Introduction: The Importance of Writing

One of the most exciting recent developments in ancient history centers around the work of Denise Schmandt-Besserat, whose theory concerning the origin of writing in Mesopotamia, the earliest known script in Western Civilization, has revolutionized our understanding of not only how writing developed but also how deep it reaches back into history. This chapter will address some of the highlights of her work, especially as it relates to the study of history. The next chapter (17) will carry that story forward through development of the alphabet.

Denise Schmandt-Besserat points out that writing is one of the great achievements of humankind, for at least three reasons. First, it represents a revolution in communication across space and time. That is, the ability to write allows our words to move far beyond the normal range of the voice and thus extends the expression of our thoughts geographically and chronologically. On this ability rests every sort of human inquiry, including history. Second, writing enables record-keeping, allowing us to study a prophet's words, engrave a tombstone or collect taxes. Crossing this threshold where the voice cannot go, the written word endures, recording the past for later review and consideration. Third, writing give us a means for scrutinizing and editing our ideas which permits us to rewrite our thoughts. As such, it opens the way to revision and greater rigor of thought, essential in logical processes of every sort, including historical investigation.

Thus, the introduction of writing marks an important crux in the history of any civilization, not only because it marks a shift in mentality toward extending communication, keeping records and re-assessing thought but also because it allows a people to live on beyond their own lives and speak to a distant future. On the written word depends every form of learning ever invented, history especially.

II. Theories of the Origins of Writing

Whether consciously or not, most people today understand how important writing is. We distinguish people as literate or illiterate, and when trying to better others' lives, one of the first things we do is teach them to read and write. Our ancient ancestors also recognized the significance of writing and many had myths recalling its invention. In Egypt, for example, the god Thoth was said to have created hieroglyphics, along with language, magic and medicine. The Mesopotamians traced the invention of writing back to Nisaba, the goddess of granaries, who they said created it to keep records of the goods coming through her temples.

Perhaps most interesting of all, at least to the modern world, is the tale preserved in Hebrew lore that Moses received the gift of writing from God along with the Ten Commandments. The Bible, after all, says explicitly that the Decalogue was "written with the finger of God" (Ex. 31:18, cf. Deut. 5.22). Israelite scholars in antiquity subsequently reasoned that God had inscribed these commandments because Moses could not write and, thus, the Hebrews must have been illiterate up until then. In this tradition, then, the Ten Commandments serve as a lesson in both morality and literacy. If a rather odd way of interpreting the Bible, this bit of folklore shows how important writing was to the ancient Israelites who made a gift from God.

The first western scholar known to have proposed a theory in which writing has a human origin was the French scholar Diderot in 1755. Based on an earlier suggestion by William Warburton, the bishop of Gloucester, Diderot suggested that early phonetic symbols developed out of pictographs, pictures representing ideas. A highly successful thesis, this proposition remained the basis for most explanations of the origin of writing in the West, until Schmandt-Besserat introduced her theory of tokens.

Much of this debate has revolved around the earliest known script in Western Civilization, cuneiform, the "wedge-shaped" system of signs used by the ancient Sumerians. Though later carved into stone, this type of writing was first impressed on clay tablets which were later fired, that is, baked so the signs won't wash away or can't be rubbed out. Finding it in deposits dating back as early as 3100 BCE, scholars theorized that cuneiform must have derived from a system of primordial pictographs. Indeed, several of its signs could be traced back to aboriginal "pictures" of the things they denoted. For example, cuneiform included a sign for "star" that looked like an asterisk, so it seemed safe to assume that it originated as some sort of depiction of a star and, therefore, all other signs derived from images, too.

But there were two major problems with postulating a pictographic origin for the great array and breadth of cuneiform signs. First, archaeological investigation has failed to produce any evidence of a forerunner for cuneiform. Instead, the physical evidence suggests this writing system arose very abruptly, seemingly out of nowhere, already in a fairly complex state. For instance, it contained right from the start at least nine hundred symbols, perhaps as many as fifteen hundred. If writing first sprang up at this moment, it came to life with a bang, almost impossibly fast.

Second, there were relatively few cuneiform signs which showed a clear lineage from pictures. The vast majority didn't look at all like what they represented, even where it would have been easy to do so. For instance, the word for "sheep" was a simple "X." Where are the legs, the wool, the horns? As early as 1928, long before Schmandt-Besserat began her work, the scholar William Mason had recognized this problem:

We must admit, that even in the earliest and most archaic inscriptions discovered, it is not always easy to recognize the original objects.

But he then went on to blame this on the ineptitude of ancient scribes:

Owing to the limitations of primitive culture, the inexperience of scribes and the lack of artistic ability, each scribe drew the characters in his own crude, faulty way, often incorrectly; so that it is quite impossible always definitely to distinguish the character and identify it with the object intended.

"Primitive" scribes with "crude" ways? Or could it be there was something wrong with theory?

Another way early historians explained these anomalies was by asserting that the Mesopotamians had deployed their now-lost system of pictographs only on biodegradable material, such as bark or animal hides. But this explanation rests on two undocumented phenomena: (1) an unknown pictographic writing system which had been executed exclusively on (2) a medium now lost. Stools with two legs missing don't make very comfortable seats. Nevertheless, in the absence of any form of writing preceding cuneiform or any better explanation for its aboriginal complexity, the pictographic theory trudged on, in spite its obvious flaws.

Over time, as more and more tablets came to light and our understanding of cuneiform improved, other issues arose to challenge further the theory of a pictographic origin. When scholars could see more clearly how early cuneiform developed, they realized that those few signs which did, in fact, arise from pictographs had been introduced after the invention of this script. That is, while the cuneiform "star" did indeed look like a star, dating suggested it was a later entry in the registry of Sumerian signs, not an early example of a type of pictography from which all cuneiform stemmed. Other historians pointed out that pictographs do not form the basis of other ancient scripts, like the Eskimo and Indian writing systems.

Another issue concerned geography. In Sumeria, the earliest cuneiform tablets come from Uruk, a major hub of civilization in the Near East and the focus of much early archaeology. But later archaeologists found evidence that cuneiform was also being used in Syria far to the west and Iran to the east at almost the same time as it first appears in Uruk. And this raised a further issue. Uruk was at that time an urbanized community with a large economy and population. Syria and Iran were relatively poor areas, sparsely populated. But cuneiform appears in all these places simultaneously. How did a complex form of writing with many abstract signs used mainly to keep a tally of properties and possessions spread with such uniformity so widely and rapidly across both city and country?

Yet another challenge to the pictograph theory came from the material it was most often written on, clay. According to the standard explanation, there's a good reason for that. There's much clay to be found in and around the Tigris and Euphrates river beds, and not much wood or animal hides. So, because of its sheer abundance alone, clay was the logical choice for Mesopotamians to use as a writing medium.

But it's not a logical choice. Clay is, in fact, very difficult to write on. First and foremost, it isn't naturally flat. It has to be pressed into a workable shape first, which is usually something rounded, something that will fit in the palm of the hand. And circular is indeed the form in which we find many a cuneiform tablet. But it's still rather difficult to write on clay even when it's carefully molded into a hand-friendly ball. Yet virtually all early cuneiform is found on clay, as if it were somehow to the ancient people of this area their traditional vehicle for writing.

In sum, the standard theory of how writing arose in Mesopotamia is full of holes and contradictions but until Schmandt-Besserat came along, there wasn't any better way to piece together the evidence. Looking at the abstract nature of even the earliest cuneiform signs, their widespread use and, most of all, the material on which they were impressed led her to a new theory and a better account of this all-important development.

III. Denise Schmandt-Besserat's New Theory of the Origin of Writing

A. Tokens

Studying Mesopotamian culture in the early 1970's, Schmandt-Besserat first set out to investigate the uses of clay before the development of pottery in early Near Eastern culture. But as she was searching for bits of clay floors, hearth linings, beads and figurines, she kept running into massive piles of small ceramic pieces found in various shapes and sizes.

At the time, these were called "enigmatic objects" or "objects of uncertain purpose," because scholars were utterly bemused about their purpose and meaning. So, for instance, when looking at a group of five cones, one archaeologist, Carleton Coon, remarked famously, "they look like nothing else in the world but suppositories. What they were used for is anyone's guess." In the process of cataloguing them as part of her research, Schmandt-Besserat first referred to them as "geometric objects" because of their configurations, until ones resembling animals and tools began to emerge. Realizing they must have stood for things, she started calling them tokens, the name by which they are now known. But still no one had any idea what they stood for, or how they were used.

Schmandt-Besserat noted some important clues, however. Many of these tokens are incised—that is, they have various markings engraved on them—and they come in a wide variety of shapes: spheres, cones, disks, cylinders and so on. Ranging in length from one to five centimeters, though they are clustered in groups of one-to-three and three-to-five centimeters, all are simple to make, in Schmandt-Besserat's words, "the shapes which emerge spontaneously when doodling with clay." That they had, in fact, been molded from wet clay, is evident from the fingerprints still preserved on some of them.

There is also a clear evolution in their design. Those found in earlier layers are plain, few in number and naturalistic in shape, while those dating to the latest times, after 3500 BCE, are more highly incised and decorated. Also there are more shapes and greater complexity among later tokens—at the same time, none required high-level skill in ceramics to create—including naturalistic renditions of beds, fruit and tools. Most intriguing of all, they stop being made after 3000 BCE, just as cuneiform enters the scene. In that final phase of their evolution, tokens revert to fewer and plainer shapes and eventually fall out of use entirely.

There are several other things notable about the nature and disposition of these tokens. For one, they come from all over the Near East: Iran, Iraq, Syria, Turkey, and Israel. For another, they date to very early times, as far back as 8000 BCE. Furthermore, there is evidence some care went into their creation because many have been fired. Firing signifies a desire to preserve them, which in turn argues that they had value of some sort. They are, in fact, among the earliest fired ceramics known.

Most of this was already evident, if inexplicable, when Schmandt-Besserat began her work. Awareness of the existence of tokens, in fact, went back almost all the way to the beginning of Near Eastern archaeology in the nineteenth century. And as early as 1959, evidence emerged that tokens represented part of a system of enumeration, functioning as counters of some sort. In particular, an envelope-tablet had been found—envelope-tablets are hollow balls of clays with tokens inside—which contained on the exterior a list of sheep and on the interior the exact number of tokens matching that inscribed on the outside. But because this was the only such tablet known, it seemed a stretch to reconstruct a entire system of token-counting based on one single piece of evidence. But, as Schmandt-Besserat later noted, the existence of many tokens having the same shape but in different sizes does, in fact, suggest they once belonged to an accounting system of some sort.

The decipherment of that system became the hallmark and triumph of her career. She noted initially that several of the designs used on and for tokens resembled later cuneiform signs. From there it was not much of a conceptual leap, though its implications to history were immense, that the tokens had originally functioned as counters representing one unit of a particular item, in much the same way later cuneiform signs denoted items in written form. But the problem wasn't really the concept, so much as its application. How did this token system of counting work, and what was it used to count? And, most important, why was it necessary?

When tokens first appeared around 8000 BCE, the vast majority of people in the world subsisted as hunter-gatherers, constantly on the move, with little or no need for counting things since nomads don't usually own much and what little they have is by necessity portable. Thus, it's surprising to find counters among the remains of civilizations dating to seventh-millennium BCE. What do they have to count? To the contrary, a settled community where goods can be stored is where one expects to find an accounting system develop, but to the first users of tokens urbanization lay far off in a distant future they could hardly have imagined.

Or so it once seemed. Recent archaeological investigations have been pushing the horizon of urbanized life back further and further in time. Settlements like Çatal Hüyük (pronounced CHAT-ul HOO-yuk) in central Turkey, which is a prehistoric community dating well back into the sixth millennium BCE, give evidence that city sites existed long before the rise of Sumerian civilization (ca. 3000 BCE). This suggests, in fact, that urbanization began at the very brink of agriculture which in some places developed as early as the eighth millennium BCE, and since farming entails a settled lifestyle and the accumulation and storage of goods, it makes sense that a counting system like tokens would also have roots that deep in history. But the need for something doesn't prove its existence. Fortunately, there's other evidence that tokens served as counters.

Similar counting systems, for one, can be found even today all over the planet. Of particular interest here, modern shepherds in Iraq still use pebbles in counting sheep. But pebbles are undifferentiated, making it unclear what they represent. That is, if a counting system employs only one type of counter, it's not possible to discriminate among various commodities. The solution to that problem is obvious and conforms precisely to the archaeological evidence seen in tokens, to differentiate the counters. Seen one way, tokens are exactly that, "differentiated pebbles."

This makes it easy to understand how tokens would have been deployed in counting, as Schmandt-Besserat argues. Say, for instance, you're a tribal chieftain and want to hold a feast. You send a runner, a young boy perhaps, off with a handful of tokens that function as a sort of "shopping list." You could also keep for yourself an identical set as a reminder of what you'd put on your "list." And you could even change your mind later and send off another boy with more tokens, in other words, a revised list. With all that, tokens clearly serve as a writing system, at least inasmuch as they are a form of communication and record-keeping in which it's possible to edit one's "words," all the hallmarks of writing.

The evolution of tokens over time only adds further to the supposition they represent an ancient accounting system of some sort. In terms of their shapes and signs, many tokens remained highly stable, changing remarkably little during their over four millennia of use in prehistory. Others, however, became more complex, especially in their latest incarnation around and after 3500 BCE as the cities of Mesopotamia were in ascendance. Their increasing vocabulary of incisions—that is, inscribed lines used to signify things—was, no doubt, the by-product of mounting urbanization. After all, as larger and more complex cities began to develop, there would have been more and more things to keep track of, necessitating a richer language of incisions to account for all that.

And one final piece of evidence attests to the use of tokens as a system of communication, the fact that many later ones have perforations, doubtlessly designed to allow them to be strung together. But why? As a filing system of some sort? Or, were they threaded on a string—a string, of course, is biodegradable and would not have survived over time—with its loose ends sealed together with a bulla, a stamped clay seal of some sort. That the more complex tokens are the ones most often found with perforations argues in favor of such an interpretation of the evidence.

B. Envelopes and Impressed Tablets

But the evidence that made Schmandt-Besserat's theory most compelling came with her study of a particular type of cuneiform document, the envelope-tablet. As noted before, it was common practice in Mesopotamian society after the invention of cuneiform to enclose a contract in a clay envelope, with a copy of the contract on the outside. This ensured no one had tampered with the details.

What Schmandt-Besserat showed was that this tradition extended far back in time, long before cuneiform itself. The envelope-tablet mentioned above in which tokens were deployed as counters had been discovered as early as 1959. Schmandt-Besserat showed this was no fluke. Other and older examples began to appear once it was clear what to look for, particularly "clay balls" with tokens inside and corresponding decorations on the outside.

Even more important, though, some of these clay balls contained the impressions of cylinder seals, long narrow stone tubes with images engraved on them in reverse so that, when they're rolled over wet clay, they leave behind a picture in relief. Because each cylinder seal is unique, Mesopotamians used them as a way of "signing" documents. Indeed, some ancient Near Eastern contracts have numerous cylinder seal impressions on them, which are, in effect, the signatures of the individuals involved in the contract.

Envelope-tablets with the text of a contract and the signatory cylinder seal impressions on the inside offered the advantage of ensuring the validity and integrity of a business transaction. But when the cuneiform document was completed and sealed inside its envelope, it was difficult to know exactly what the contract stipulated since the clay envelope hid the text inside. Archaeology shows, however, that the ancients found a ready solution to that problem. They copied the contract onto the envelope itself.

Schmandt-Besserat's contribution, arguably one of her greatest, was to show how old this practice really was, that in its earliest manifestation the envelope-tablet didn't utilize cuneiform writing but tokens themselves pushed into the wet clay of an envelope which left their impression on it. While the clay is still wet, the tokens themselves were sealed inside, and the whole package was left to dry or be fired. The copy of the tokens on the envelope is itself an important conceptual leap, a first step toward representing tokens abstractly as two-dimensional cuneiform signs, not three-dimensional tokens.

The next step was to stop impressing the tokens on the envelope and instead draw their picture on the envelope's wet clay, an advancement which followed soon thereafter. This was especially necessary for incised tokens, because their marks which are crucial to their meaning do not transfer well onto wet clay. And as incised tokens became more popular in the economic boom starting around 3500 BCE, the need to represent them precisely on envelopes would only have increased.

Finally the ancient Mesopotamians must have realized that, if the tokens inside are represented on the envelope and the tablet is fired making it impossible to alter it in any way, the tokens themselves inside the envelope aren't necessary. All a contract really needed to ensure its lasting validity was the symbolic signs on the outer envelope, originally an exterior copy of the contract but now the whole contract itself. With this, it makes sense that cuneiform signs derived from the shapes and markings on tokens, which do, indeed, constitute a "picture" of sorts but not the sort of picture expected in the standard view of a pictograph. It's a picture of a token, not a picture of the thing itself. Schmandt-Besserat sums it up this way:

[T]hese types of symbols, which derived from tokens, were picture signs or "pictographs." They were not, however, pictographs of the kind anticipated by Warburton. The signs were not pictures of the items they represented but, rather, pictures of the tokens used as counters in the previous accounting system. . . The first tablets were a decisive step in the invention of writing and amounted to a revolution in communication technology.

The last step to a full and independent writing system was the creation of new cuneiform signs not based on an original token design. That must first have happened when the thing being represented wasn't a commodity of exchange at all in the traditional token economy, no "sheep" or "bar of metal," but something like "star" or "man." And this explains why true pictographic cuneiform signs date, for the most part, to later, not earlier times. It also clarifies why many are expressed as a rebus, a "word-picture" in which the elements of the word are written as separate and unrelated pictures, like writing "carpet" by drawing an automobile and a cat or dog.

From there, it's no long trek to creating a syllabary, that is, a writing system representing spoken syllables, in which any word can be spelled phonetically. As we'll see in Chapter 17, the Western alphabet arose from just that. Thus, Denise Schmandt-Besserat's innovative researches have brilliantly elucidated the origins of writing, which we now know evolved from distinctive clay tokens to envelope-tablets with impressions of tokens on the outside to token-free tablets written in cuneiform, eventually containing its own complex syllabary.

Besides that, it also shed light on two idiosyncrasies of Mesopotamian writing: why the Mesopotamians wrote on clay as opposed to some more convenient—or, at least, naturally flatter—medium, and why many cuneiform tablets are round. Ever since the earliest use of tokens, clay was the traditional medium of accounting transactions in Mesopotamia. And not just because it was convenient to hold clay in a ball did cuneiform tablets tend to come in a rounded form, but because it was also the traditional shape used for early clay envelopes enclosing tokens. Much like changing people's minds today about writing on paper, "hard copy" that is, or using a size of paper other than eight-and-a-half-by-eleven-inches, cultural traditions can be deeply entrenched, even after technical advancements make them obsolete.

IV. Conclusion: The Origin of Writing in Tokens

Thus, as a system of communication, tokens fulfill the three fundamental purposes which make writing the monumental invention it is. First, the use of tokens allowed ancient peoples to communicate across unprecedented expanses of space and time far exceeding the range of the voice. Second, they constituted a form of record-keeping, permitting the precise determination of how many there were of something, and even to distinguish between different types of item. And finally, the ability to re-send messages, to renegotiate contracts, to string tokens together and then re-string them entails perhaps the most important feature of this writing system, the revision process which admits scrutiny and editing of thought.

Perhaps most important of all to note, the discovery of all this came not from digging up new information but the study of what had already been uncovered and was sitting in museums, and in some cases had been for decades. It took fresh insight, a newcomer's eyes, to realize all that it represented. Little wonder, then, Schmandt-Besserat's theory was nominated as among the top one-hundred scientific theories of the twentieth century. It illuminates, after all, one of the top twenty inventions in a hundred centuries.