The New York Times -- April 6, 1999

Who Began Writing? Many Theories, Few Answers

By JOHN NOBLE WILFORD

HILADELPHIA -- The Sumerians had a story to explain their invention of writing more than 5,000 years ago. It seems a messenger of the king of Uruk arrived at the court of a distant ruler so exhausted from the journey that he was unable to deliver the oral message. So the king, being clever, came up with a solution. He patted some clay and set down the words of his next messages on a tablet.

A Sumerian epic celebrates the achievement:

Before that time writing on clay had not yet existed,
But now, as the sun rose, so it was!
The king of Kullaba [Uruk] had set words on a tablet, so it was!

A charming just-so, or so-it-was, story, its retelling at a recent symposium on the origins or writing, held here at the University of Pennsylvania, both amused and frustrated scholars. It reminded them that they could expect little help -- only a myth -- from the Sumerians themselves, presumably the first writing people, in understanding how and why the invention responsible for the great divide in human culture between prehistory and history had come about.

	Early Writing: Cultures Make Their First Marks
	Representing spoken language in writing grew out of the transformation of pictures and random marks into images with consistent meaning. Among the earliest cultures to do this were:
	SUMERIA, 3,200 B.C. This Mesopotamian culture is generally thought to be the first culture to produce written texts. In the example above the Sumerians used a stylus and wet clay to record the ingredients for beer.
	EGYPT, 3,200 - 3,000 B.C. Early Egyptians used a system of pictorial hiero-glyphics as a written language. The earliest examples (above) were phonetically arranged symbols on small clay tablets used to indicate payments for commodities. It reads “Mountain of Light.”
	INDUS VALLEY, 2,500 B.C. Ceramic seals like the one shown above were used by the inhabitants of what is now Pakistan to indicate ownership or destinations for bundles of goods. The scripts in the upper right corner are individual characters.
	CHINA, 1,500 - 1,200 B.C. During the Shang Dynasty the Chinese began to inscribe divinations on the underside of turtle shells and ox bones. These artifacts are now referred to as oracle bone inscriptions.
	MESOAMERICA 250 - 300 B.C. The Olmecs, Mayans and Zapotecs are thought to be the first of many cultures in this region to develop writing. Shown above is an early Zapotec example that is still untranslated. Sources: “Old World Civilizations,” by the American Museum of Natural History Jim McManus/The New York Times; Photographs courtesy of Margaret Nissen, German Archeological Institute and Harappa Archeological Research Project

The archeologists, historians and other scholars at the meeting smiled at the absurdity of a king's writing a letter that its recipient could not read. They also doubted that the earliest writing was a direct rendering of speech. Writing more than likely began as a separate and distinct symbolic system of communication, like painting, sculpture and oral storytelling, and only later merged with spoken language.

Yet in the story, the Sumerians, who lived in Mesopotamia, the lower valley of the Tigris and Euphrates Rivers in what is now southern Iraq, seemed to understand writing's transforming function. As Dr. Holly Pittman, director of the university's Center for Ancient Studies and organizer of the symposium, observed, writing "arose out of the need to store information and transmit information outside of human memory and over time and over space."

In exchanging interpretations and new information, the scholars acknowledged that they still had no fully satisfying answers to the most important questions of exactly how and why writing was developed. Many of them favored a broad explanation of writing's origins in the visual arts, pictograms of things being transformed into increasingly abstract symbols for things, names and eventually words in speech. Their views clashed with a widely held theory among archeologists that writing grew out of the pieces of clay in assorted sizes and shapes that Sumerian accountants had used as tokens to keep track of livestock and stores of grain.

The scholars at the meeting also conceded that they had no definitive answer to the question of whether writing was invented only once and spread elsewhere or arose independently several times in several places, like Egypt, the Indus Valley, China and among the Olmecs and Maya of Mexico and Central America. But they criticized recent findings suggesting that writing might have developed earlier in Egypt than in Mesopotamia.

In December, Dr. Günter Dreyer, director of the German Archeological Institute in Egypt, announced new radiocarbon dates for tombs at Abydos, on the Nile about 250 miles south of Cairo. The dates indicated that some hieroglyphic inscriptions on pots, bone and ivory in the tombs were made at least as early as 3200 B.C., possibly 3400. It was now an "open question," Dr. Dreyer said, whether writing appeared first in Egypt or Mesopotamia.

At the symposium, Dr. John Baines, an Oxford University Egyptologist who had just visited Dr. Dreyer, expressed skepticism in polite terms. "I'm suspicious of the dates," he said in an interview. "I think he's being very bold in his readings of these things."

The preponderance of archeological evidence has shown that the urbanizing Sumerians were the first to develop writing, in 3200 or 3300 B.C. These are the dates for many clay tablets with a proto-cuneiform script found at the site of the ancient city of Uruk. The tablets bore pictorial symbols for the names of people, places and things for governing and commerce. The Sumerian script gradually evolved from the pictorial to the abstract, but it was probably at least five centuries before the writing came to represent recorded spoken language.

Egyptian hieroglyphics are so different from Sumerian cuneiform, Dr. Baines said, that they were probably invented independently not long after Sumerian writing. If anything, the Egyptians may have gotten the idea of writing from the Sumerians, with whom they had contacts in Syria, but nothing more.

In any event, the writing idea became more widespread at the beginning of the third millennium B.C. The Elamites of southern Iran developed a proto-writing system then, perhaps influenced by the proto-cuneiform of their Sumerian neighbors, and before the millennium was out, writing appeared in the Indus River Valley of what is now Pakistan and western India, then in Syria and Crete and parts of Turkey. Writing in China dates back to the Shang period toward the end of the second millennium B.C., and it dates to the first millennium B.C. in Mesoamerica.

Archeologists have thought that the undeciphered Indus script, which seemed to appear first around 2500, may have been inspired in part from trade contacts with Mesopotamia. But new excavations in the ruins of the ancient city of Harappa suggest an earlier and presumably independent origin of Indus writing.

In a report from the field, distributed on the Internet, Dr. Jonathan Mark Kenoyer of the University of Wisconsin and Dr. Richard H. Meadow of Harvard University showed pictures of marks incised on potshards that they interpreted as evidence for the use of writing signs by Indus people as early as 3300 B.C. If these are indeed proto-writing examples, the discovery indicates an independent origin of Indus writing contemporary with the Sumerian and Egyptian inventions.

Dr. Meadow, using E-mail, the electronic age's version of the king of Uruk's clay tablet, confirmed that the inscribed marks were "similar in some respects to those later used in the Indus script." The current excavations, he added, were uncovering "very significant findings at Harappa with respect to the Indus script."

At the symposium, though, Dr. Gregory L. Possehl, a Pennsylvania archeologist who specializes in the Indus civilization and had examined the pictures, cautioned against jumping to such conclusions. One had to be careful, he said, not to confuse potter's marks, graffiti and fingernail marks with symbols of nascent writing.

Of the earliest writing systems, scholars said, only the Sumerian, Chinese and Mesoamerican ones seemed clearly to be independent inventions. Reviewing the relationship between early Chinese bronze art, "oracle bones" and writing, Dr. Louisa Huber, a researcher at Harvard's Fairbanks Center for East Asian Research, concluded, "Chinese writing looks to be pristine."

But few pronouncements about early writing go undisputed. Dr. Victor Mair, a professor of Chinese language at Penn, offered evidence indicating, he said, that "the Chinese writing system may well have received vital inputs from West Asian and European systems of writing and proto-writing."

Dr. Mair cited an intriguing correspondence between the Chinese script and 22 Phoenician letters and also Western-like symbols on pottery and the bodies of mummies found in the western desert of China. The recent discoveries of the mummies, wearing garments of Western weaves and having Caucasoid facial features, have prompted theories of foreign influences on Chinese culture in the first and second millennia B.C. It had already been established that the chariot and bronze metallurgy reached China from the West.

Though no one seemed ready to endorse his thesis, Dr. Mair said, "We simply do not know for certain whether the Chinese script was or was not independently created."

Dr. Peter Damerow, a specialist in Sumerian cuneiform at the Max Planck Institute for the History of Science in Berlin, said, "Whatever the mutual influences of writing systems of different cultures may be, their great variety shows, at least, that the development of writing, once it is initiated, attains a considerable degree of independence and flexibility to adapt a coding system to specific characteristics of the language to be represented."

Not that he accepted the conventional view that writing necessarily started as some kind of representation of words by pictures. New studies of Sumerian proto-cuneiform, he said, challenge this intepretation. The structures of this earliest writing, for example, did not match the syntax of a language. Proto-cuneiform seemed severely restricted, compared with spoken language, dealing mainly in lists and categories, not in sentences and narrative.

This presumably reflects writing's origins and first applications in economic administration in a growing, increasingly complex society, scholars said. Most of the Uruk tablets were documents about property, inventory and, even then, taxes. The only texts that do not concern administrative activities, Dr. Damerow said, were cuneiform lexicons that were apparently written as school exercises by scribes in training.

For at least two decades, in fact, Dr. Denise Schmandt-Besserat, a University of Texas archeologist, has argued that the first writing grew directly out of a counting system practiced by Sumerian accountants. They used molded clay "tokens," each one specially shaped to represent a jar of oil, a large or small container of grain, or a particular kind of livestock. When the tokens were placed inside hollow clay spheres, the number and type of tokens inside was recorded on the ball with impressions resembling the tokens. Finally, simplifying matters, the token impressions were replaced with inscribed signs, and writing was invented.

Though Dr. Schmandt-Besserat has won wide support, some linguists question her thesis and other scholars, like Dr. Pittman of Penn, think it too narrow an interpretation. They emphasized that pictorial representation and writing evolved together, part of the same cultural context that fostered experimentation in communication through symbols.

"There's no question that the token system is a forerunner of writing, and really important," Dr. Pittman said in an interview. "But I have an argument with her evidence for a link between tokens and signs, and she doesn't open up the process to include picture-making and all other kinds of information-storage practices that are as important as the tokens."

Dr. Schmandt-Besserat, who did not attend the symposium, vigorously defended herself in a telephone interview. "My colleagues say the signs on seals were a beginning of writing, but show me a single sign on a seal that becomes a sign in writing," she said. "They say that designs on pottery were a beginning of writing, but show me a single sign of writing you can trace back to a pot -- it doesn't exist."

In its first 500 years, she asserted, cuneiform writing was used almost solely for recording economic information. "The first information that writing gives you is only the same information the tokens were dealing with," she said. "When you start putting more on the tablets, products plus the name of who has delivered and received them, that is where art would enter the picture. Then writing is out of the box, in all directions."

Dr. Damerow agreed that cuneiform writing appeared to have developed in two stages, first as a new but limited means of recording economic information, later as a broader encoding of spoken language for stories, arguments, descriptions or messages from one ruler to another.

Even so, it was a long way from the origin of writing to truly literate societies. At the symposium, scholars noted that the early rulers could not write or read; they relied on scribes for their messages, record keeping and storytelling. In Egypt, most early hieroglyphics were inscribed in places beyond the public eye, high on monuments or deep in tombs.

In this case, said Dr. Pascal Vernus of the University of Paris, early writing was less administrative than sacred and ideological, "a way of creating and describing the world as a dominating elite wants it to be."

Dr. Piotr Michalowski, professor of Near East civilizations at the University of Michigan, said the Uruk proto-cuneiform writing, whatever its antecedents, was "so radically different as to be a complete break with the past, a system different from anything else." It no doubt served to store, preserve and communicate information, but also was a new instrument of power.

"Perhaps it's because I grew up in Stalinist Poland," Dr. Michalowski said, "but I say coercion and control were early writing's first important purpose, a new way to control how people live."