You are here

The First Writing: Script Invention as History and Process

July 2006 (110.3)

Book Review

The First Writing: Script Invention as History and Process

Edited by Stephen D. Houston. Pp. xvii + 417, figs. 107, tables 5, maps 3. Cambridge University Press, Cambridge 2004. $90. ISBN 0-521-83861-4 (cloth).

Reviewed by

Writing was invented, that is a fact of history. Was it invented more than once, or was it diffused from one source? Why was it invented? For royal display, for religious purposes, or for banal administration? How was the original invention developed, and when did it come to represent speech adequately? These are some of the questions investigated by the 12 essays in this volume, the product of a colloquium organized by the editor in 2003, provoked by major advances in understanding Mesoamerican scripts.

Stephen Houston surveys “Writing in Early Mesoamerica” (274–309), showing the characteristics of writing that are now indisputably evident (pace Gelb). From the mid first millennium B.C. there are specimens from several areas displaying progress beyond glyphic markers of identity to representations of language, although decipherment is limited. The slightly later examples of Mayan script, from about 2,000 years ago, imply that religion was its primary context with most writing now lost, done with ink or paint on perishable surfaces. That lack makes it impossible to argue that all writing originated in administration, as Houston observes in his “Final Thoughts on First Writing” (349–53). Mayan script was as able to mark all parts of speech as the Chinese, cuneiform, or Egyptian scripts.

The authors are conscious that other scripts inadequately represent the ways they were used at the start. Robert W. Bagley demonstrates this most plainly in “Anyang Writing and the Origin of the Chinese Writing System” (190–249). While inscribed “oracle bones” and bronze vessels survive in the thousands from the Shang Dynasty (ca. 1200 B.C.), the sign for “document” that occurs on them is a set of bamboo strips. Examples of those survive only from about 700 years later; their perishability has robbed us, so far, of the earlier stages that led to the connected discourse that oracle bones present. Bagley lucidly discusses the Chinese evidence and hypothesizes about the rise of writing in China, which, he believes, was an independent invention. In contrast, Françoise Bottéro (“Writing on Shell and Bone in Shang China,” 250–61) sees the absence of any primary stage of Chinese writing as evidence for the diffusion of the idea of writing from the West. Despite the lack of material, however, Bagley compares the Chinese situation with the Mesopotamian and Egyptian and argues for a primary context in administrative record keeping.

That is clear in Mesopotamia, as Jerrold S. Cooper explains in “Babylonian Beginnings: The Origin of the Cuneiform Writing System in Comparative Perspective” (71–99). Newly available material and the work of a major research project on the earliest texts (ca. 3200 B.C.) expand our knowledge of the system, initially logographic but rapidly adopting the rebus principle to produce phonetic writing, already present but rare in the oldest texts. The Cuneiform Digital Library Initiative at UCLA is one of those research projects, and Robert K. Englund’s “The State of Decipherment of Proto-Elamite” (100–49) is a fruit of its work. A corpus of 1,600 texts written on clay tablets in Iran about 3000 B.C. remains unread. Englund sets out typical texts, interprets the complex numerical systems, shows the debt to the Babylonian script, and demonstrates how electronic analysis can map the uses of one sign. Yet prospects for full decipherment remain remote.

The situation is similar in Egypt, where German excavators of a royal tomb at Abydos have recently uncovered numerous examples of a very early form of Egyptian writing. Bone tags about 1.5 cm high bear one to five incised signs and numerals, while pottery jars bear one or two painted signs, patently ancestral to the hieroglyphs. Similar tags had been found a century ago, but now their dating is clear: the Naqada III period, ca. 3200 B.C. The tags were evidently tied to consignments of goods now lost, probably giving names of estates or their owners, according to John Baines (“The Earliest Egyptian Writing: Development, Context, Purpose,” 150–89). Baines contrasts the signs on these small tablets with the signs on the jars, which occupy up to half the height of the vessels. He concludes of those, “their scale suggests that they constitute some sort of display and do not only convey information” (158). He goes on to discuss the place of hieroglyphs in early Egyptian art; while he admits that the Abydos “material appears to be administrative,” he is uncertain about the value of marking provenances on goods to be buried in a tomb (170). Here it may be proposed that the slightly later use of hieroglyphic signs in conjunction with figural representations (e.g., on the Hunters’ palette and the Scorpion mace-head, illustrated 168–69) is a misleading analogy. If the tomb’s contents, placed there for use by the deceased’s spirit, mirrored the provisions of the palace in the king’s lifetime, then the hundreds of jars in his tomb duplicate the palace stores, some of them marked for easy identification rather than display. Therefore, even in the absence of archaic texts from occupied sites, it seems justified to argue, against Baines, that the purpose of writing in Egypt was primarily administrative and its symbolic aspect was less significant.

Bruce Trigger’s “Writing Systems: A Case Study in Cultural Evolution” (39–68) surveys semasiographic (pictures or ideograms), logophonic (signs for words and sounds), and phonographic (sounds only) systems, showing how the first gave rise to logophonic systems in Sumer, Egypt, China, and Mesoamerica, with phonographic ones developing from knowledge of logophonic, although not evolving directly from them. Phonographic systems often appear when scripts were adapted or invented for other languages. However, there “are no clear examples of a functioning syllabic script developing directly from a logophonic one” (55); the term “syllabic” covers syllabaries, consonantaries, alphabets, and alphasyllabaries (where each sign marks a consonant plus specific vowel). For Trigger, the development of writing systems may offer “a model for trying to understand changes in other forms of culture,” displaying tendencies such as efficiency and social interests (67).

A moment of decisive change in script invention was the production of the alphabet. Regrettably, no chapter is devoted to that, although it has a significant place in Trigger’s paper. Here is an invention that had to be made at one moment, complete (or almost so). By alphabet I mean the script ancestral to the Phoenician, a system of writing that is now called a “consonantary” (i.e., it notes consonants only). Whereas pictographic, logographic, and even syllabic scripts could develop by accretion, the alphabet had to be created complete. Whatever pattern may, or may not, have existed in the Egyptian “hieroglyphic alphabet” (never isolated as such until very late in Egyptian history), the inventor of the alphabet had to make a comprehensive analysis of his language to ensure that he created a distinct sign for each of its major consonantal phonemes. Further research is needed into this major step and the relevance of cuneiform syllabaries, which indicate awareness of the Sumerian phonetic stock. The only essay to deal with an alphabetic script is Henrik William’s “Reasons for Runes” (262–73). He considers the earliest texts (from about A.D. 200 onward), which are mostly names, and argues that the script was invented on the basis of the Latin alphabet for the elite and had little practical function. This is a different category of script, as it is obviously derivative and deserves to be compared with others of the same sort (e.g., Iberian). The possibility that the absence of longer runic texts from the early period is due to perishable writing materials should not be wholly discounted.

Writing is not the only means of communication in permanent, visible form, as Elizabeth Boone demonstrates (“Beyond Writing,” 313–48). She offers examples of “non-phonetic graphic systems,” such as notations used in mathematics and physics, music and dance, diagrams in physics and chemistry, and tables and charts of many kinds. All of these, in fact, are better described as adjuncts of writing; they would not be intelligible to the uninitiated without texts. The Mexican codices, her special study, are diagrammatic, like some ancient Egyptian tomb paintings, serving as prompts, giving information visually, which might then be verbalized. They are thus a form of pictography.

“The Possibility and Actuality of Writing” (16–38) is the title of John Robertson’s opening essay. Taking Peirce’s icon-index-symbol triad as his tool, he works through the relationship of writing to speech, the reason why pictographic symbols give rise to ideographs and why phonography may follow. Writing reflects speech, so some of the laws governing language apply to writing, enough to ensure adequate representation, something that tends to grow as scripts develop.

This collection of papers should serve as a stimulus to further comparative study of the reasons for writing and the different stages of script development. Two observations are notable: writing is not a necessary characteristic of civilization (40), and scripts flourish as centers collapse (76, 293). The universality of the rebus principle in logophonic scripts reveals an intriguing aspect of the human mind embedded in one of its most significant inventions.

Alan Millard
School of Archaeology, Classics and Egyptology
14, Abercromby Square
University of Liverpool
Liverpool L69 7WZ
United Kingdom

Book Review of The First Writing: Script Invention as History and Process, edited by Stephen D. Houston

Reviewed by Alan Millard

American Journal of Archaeology Vol. 110, No. 3 (July 2006)

Published online at

DOI: 10.3764/ajaonline1103.Millard

Add new comment

Plain text

  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.