Contents
Cover
Half Title page
Title page
Copyright page
Dedication
Illustrations
Maps
Preface
Chronology
Introduction: A Difficult Topic, Little Studied, Poorly Understood
A Chaos of Terms
Writing, Language, Speech
Chapter 1: What Is Writing?
The Magic, Romance, and Danger of Writing
A Definition of Writing
Change and Evolution in Systems of Writing
Writing Is Material
Speech and Writing
Chapter 2: Writing with Signs
Gelb’s Category “Forerunners of Writing”
Semasiography
Observations
Chapter 3: Categories and Features of Writing
The Rebus
Logography, the First Division of Lexigraphy
Phonography, the Second Division of Lexigraphy
Syllabography and Alphabetic Writing, the Two Categories of Phonography
Auxiliary Signs and Devices
Spelling Rules
Orthography
Chapter 4: Some General Issues in the Study of Writing
Strategies in the Formation of Lexigraphic Writing Systems
Writing and Play
“You Have a Lovely Hand”: Writing and Beauty
Chapter 5: Protocuneiform and Counting Tokens
The Protocuneiform Tablets of Uruk III and IV
Context for Protocuneiform Writing
The Tokens of Denise Schmandt-Besserat
Chapter 6: Origin of Lexigraphic Writing in Mesopotamia
Discovery of the Phonetic Principle
The Discovery and Decipherment of Cuneiform
Logosyllabic Cuneiform Writing
Transliteration Nightmares
Changes Across Time and Place
Summary
Chapter 7: Plato’s Ideas and Champollion’s Decipherment of the Egyptian Hieroglyphs
The Allegorical Interpretation of the Hieroglyphs
Background to the Decipherment of the Hieroglyphs
The Decipherment of Jean François Champollion
Chapter 8: Egyptian Writing and Egyptian Speech
The Phases of Egyptian Language/Speech
The Forms of Egyptian Writing
Chapter 9: The Origin and Nature of Egyptian Writing
The Earliest Egyptian Writing
Different Kinds of Signs in Egyptian Writing
Types of Phonograms
Nonphonetic Signs: Logograms and Semantic Complements/Determinatives
Chapter 10: “The House of Life”: Scribes and Writing in Ancient Egypt
The Tools of the Egyptian Scribe
The Role of Scribes
An Example of Egyptian Writing
Chapter 11: Syllabic Scripts of the Aegean
“Cretan hieroglyphs”
Linear A
Linear B
The Decipherment of Linear B
How Linear B Works
Syllabic Writing on Cyprus
Chapter 12: The West Semitic Revolution
Cuneiform Syllabaries
The West Semitic Syllabaries
The “Ugaritic Cuneiform Alphabet”
The Phoenician Syllabary, c.1000 BC
Chapter 13: What Kind of Writing Was West Semitic?
What Is an Alphabet?
The Phoneme
The Phoneme as a Projection of Greek Alphabetic Writing
Abjads, Abugidas, and Other Monsters
Chapter 14: The Origins of West Semitic Writing
Origins of West Semitic Writing: The Epigraphic Finds
The Proto-Sinaitic Inscriptions
Discarding the “Acrophonic Principle” in the History of Writing
The Invention of West Semitic Writing
Other Levantine Epigraphic Finds from the Bronze Age
Chapter 15: Chinese Logography
Chinese Writing
Neolithic Finds
Oracle Bones and the Problem of Origins
How Chinese Writing Works
An Example of a Chinese Complex Character
Chinese Logography
Chinese Writing and Chinese Speech
Attempts at Reform
China’s Influence
Chinese Writing and Poetic Expression
Summary
Chapter 16: Lexigraphic Writing in Mesoamerica
Slouching toward Decipherment
Origins of Maya Writing
The Nature of Maya Writing
The Earliest Historical Text
Translating a Mayan Text: A Historical Inscription from DOS PILAS, Mexico
Summary
Chapter 17: The Greek Alphabet: A Writing That Changed the World
Background to the Invention of the Greek Alphabet
The Adapter’s Achievement
The Earliest Greek Alphabetic Writing
The Date of the Alphabet’s Invention
The Poetic Inspiration for the Invention of the Alphabet
The Fortuitous Origins of Alphabetic Writing
Chapter 18: Summary and Conclusions
Glossary
Bibliography
Further Reading
References
Index
This paperback edition first published 2012
© Barry B. Powell
Edition history: Blackwell Publishing Ltd (hardback, 2009)
Blackwell Publishing was acquired by John Wiley & Sons in February 2007. Blackwell’s publishing program has been merged with Wiley’s global Scientific, Technical, and Medical business to form Wiley-Blackwell.
Registered Office
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom
Editorial Offices
350 Main Street, Malden, MA 02148-5020, USA
9600 Garsington Road, Oxford, OX4 2DQ, UK
The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
For details of our global editorial offices, for customer services, and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.
The right of Barry B. Powell to be identified as the author of this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data
Powell, Barry B.
Writing : theory and history of the technology of civilization / by Barry B. Powell.
p. cm.
Includes bibliographical references and index.
ISBN 978-1-4051-6256-2 (hardcover : alk. paper) ISBN 978-1-118-25532-2 (pbk. : alk. paper)
1. Writing–History. 2. Writing–Social aspects. [1. Alphabet–History.] I. Title.
P211.P69 2009
411.09–dc22
2008046991
A catalogue record for this book is available from the British Library.
Preface
I hope this book may serve as a brief introduction to an immense, tangled, and obscure topic. Writing can be defined and understood, but only with the help of a careful organization of categories and terms. I know of no other humanistic topic more distorted through the careless use of categories and terms, so that things “everyone knows” are illusions. The professionals, too, offer us neologisms, buzzwords, and terms that attempt a fatal precision. For example, in one of the best books on writing in the last several years (S. D. Houston, ed., The First Writing: Script Invention as History and Process, Cambridge, 2004), a series of essays on the origins of writing, the reader will struggle with “glottography,” “cipherability,” “morphophonic,” “alphasyllabaries,” “consonantaries,” “logophonic,” “logophonemic,” “logoconsonantal,” “phonological heterography,” “taxograms,” “semasiologographic,” “graphotactical,” “numero-ideographic,” “phonophoric,” “ethnogenetic” – as well as the usual bête noire “pictograms” and “ideograms.” Is writing really so complex, or esoteric? The study of the history of writing is the study of the explosion of illusions, and such jargon has stood as the greatest obstacle to understanding. Yet we cannot understand the historical past without understanding the technology that made possible our knowledge of it. This book should be of interest to anyone who wishes to come to grips with the question, What happened in the human past?
I have dedicated this book to my friend and colleague Emmett L. Bennett, Jr., but would also like to thank him here for the countless insights into the history of writing he has given me over the years, and the collection of examples that illustrate these insights.
I should also like to thank John Bennet for reading the entire manuscript and saving me from many errors, both of fact and interpretation. I am deeply grateful to him for his help. His hand appears on nearly every page, but I reserve to my own responsibility all remaining failings of both kinds.
To annotate a book such as this properly would require massive documentation that would detract from the synthesis I propose. I have therefore reserved remarks about bibliography to a section in the back.
Photos and translations not otherwise credited are my own. I have included basic maps with many chapters, because where things happened is as important as when. In the text I have highlighted places on the maps by means of small capitals. The reader may find the glossary at the back of the book useful in keeping straight the bewildering terminology of writing.
Introduction: A Difficult Topic, Little Studied, Poorly Understood
It is not hard to see that writing is the single most important technology in human life, yet it is not easy to study or to think about. Nonetheless we use it almost every minute of our lives. Naturally, many handbooks attempt to explain this extraordinary technology, some of good quality, but most suffer from a recurring blindness about what writing is, where it comes from, and how it functions in relationship to speech. All scientific speculation on the history of writing, without exception, is conducted by alphabet-users, including the present study, which gives a bias to our questions and to what we take as answers. Many historians of writing do not read nonalphabetic scripts or have a casual acquaintance with them. The alphabet-using historians of writing make prejudgments that harm our understanding.
In this book I will struggle against such prejudgments by providing a scientific nomenclature for understanding writing built on a coherent model of the different internal structures that govern all writing. I want to explicate this nomenclature and this model (see diagram facing the title page) through the study of the history of writing in the ancient Mediterranean, China, and Mesoamerica. This book is not, then, a description of the endless variety of external form in the history of writing, for which good studies exist, but an examination through historical examples of the internal structural principles that govern all writing. By proceeding in this fashion through a dark forest filled with dragons, I hope to slay several and clear away some popular confusions:
- the illusion that the purpose, origin, and function of writing is to represent speech
- the common supposition that writing comes from pictures
- the misapprehension that writing necessarily evolves toward the goal of finer phonetic representation
A Chaos of Terms
In no sense is the history of writing a discipline with niches in universities filled by experts. Those who write about the history of writing come from different directions and bring with them the expectations of their own disciplines. Linguists occasionally write such books because they feel that “language” is their province and that writing is somehow language. They are unrealistic about the quality of the phonetic information encoded in systems of writing, and their explanations too often ignore the social and historical forces behind change in systems of writing. Archaeologists sometimes work directly with unfamiliar ancient scripts, but they are rarely trained philologists. Perhaps philologists are in the best position to study the history of writing, if they have learned a nonalphabetic script, because they have wrestled most with the problem of deriving meaning from symbols. Thus the Polish-American Assyriologist I. J. Gelb (1907–85), who worked at the University of Chicago and contributed to the decipherment of “Luvian hieroglyphs,” wrote the most important analysis of writing in the twentieth century and laid the foundations for the modern scientific study of writing. His famous book A Study of Writing: The Foundations of Grammatology appeared first in 1952 (revised in 1963). Gelb was wrong in details, but he understood that the history of writing exists, with discoverable underlying principles, as in all historical study. His outline of those principles stands today, and I refer to them often in this book.
Above all Gelb urged the use of a consistent and rational vocabulary in discussing the history of writing, although few follow his advice. In reading and thinking about writing we struggle with terms that have their origin in the history of study, not in the nature of the subject. For example, we just referred to “Luvian hieroglyphs” to distinguish this writing from “Hittite cuneiform,” but there is nothing hieroglyphic about this writing except the casual and entirely superficial resemblance to the historically unrelated Egyptian hieroglyphs. Both scripts are iconic, that is, we can sometimes recognize in the signs objects from the everyday world – for example, a hand, a bird, or an animal – but there is no direct historical connection between the scripts and they work in different ways. Another example is the “Ugaritic cuneiform alphabet,” which is unrelated to Mesopotamian cuneiform and is not an alphabet. Unfortunately, such terms have stuck, and we are stuck with them, and we are stuck constantly explaining that this or that term is inappropriate. I will put such casual and inaccurate but common terms in “quotations.”
But the misuse of three words more than any others have harmed the study of the history of writing: “pictogram,” “ideogram” (or ideograph), and “alphabet.” The word “pictogram” means “picture-writing,” but carries with it so much imprecision that we must avoid it rigorously. The use of “pictogram” should be the hallmark of the amateur, but careless professionals go on using it. It is always tempting to call any sign that looks like something a “pictogram,” implying that the message is communicated through pictures and not through the resources of speech. Underlying the use but usually unspoken is a specious theory that “writing” began as pictures, then somehow became attached to speech, yet still remained pictures. So written characters that resemble something in the world, like Egyptian hieroglyphs, are called pictograms, as if the sign stands for what they picture and not for elements of human speech or sounds of speech. We do find representations of recognizable objects in early writing, but these “pictures” can fulfill a wide range of functions. Even when such designs appear to stand for the object represented, really they stand for the word attached to the object; that is, they refer to elements of speech, and not directly to items found in the world. “Pictures” can fulfill other functions, for example, place the thing described in a category. When wishing to speak of the representational aspects of some writings, we can call these aspects “iconic”.
A similar situation pertains to the word “ideogram,” often used, for example, of Chinese characters or of a class of signs in Cretan Linear B. Ideogram should mean “idea-writing,” that is, the graphic symbolization of an “idea,” a Platonic, invisible, eternal, unspoken reality. At one time scholars thought that Egyptian hieroglyphs were just that. But Chinese characters never represent eternal, unspoken realities, and neither do Linear B characters. Probably “ideograms” do not exist, so it makes little sense to talk about them.
The deep problems surrounding the word “alphabet” will be the subject of a good part of this book.
Writing, Language, Speech
Writing is old, but writing attached to speech, which I will call lexigraphic writing, goes back only to around 3400 BC, as far as we know. That is a rough date, as are all dates in the fourth millennium, but one we can nonetheless work with. Immense changes have taken place in the art of lexigraphic writing since that time, as one will quickly discover if setting out to learn Akkadian cuneiform. Yet such changes rarely result from evolution, except in writing’s earliest stages, and were never inevitable. They came about through the accidents of history and the intercession of individual men of genius working across racial and linguistic bounds, when fresh approaches were possible. There is no certain direction that a writing must take. Because writing systems are arbitrary and conventional they do not respond to nature (whose rules of behavior are not arbitrary and not conventional), but to the inventiveness of unknown creators, who had a purpose too often hidden from us.
So improbable is it that anyone should devise a means of encoding elements of speech by means of graphic symbols that in the Old World lexigraphic writing was invented only once, in Mesopotamia, and perhaps a second time, much later, in China. But even in China the idea of “writing” must have come from Mesopotamia over the Gansu corridor north of the Himalayas, where caravan traffic was constant. China was never wholly separated from cultural developments in Mesopotamia. A separate invention did take place in Mesoamerica, providing a test case for principles distilled from the study of Near Eastern writing. We will spend one chapter on writing not attached to speech, which I call semasiography (after I. J. Gelb), but most of this book is about lexigraphic writing.
Because such writing is attached to speech, we need a clear description of what we mean by speech. Unfortunately scholars often use “language” when really they mean “speech,” as if they were the same thing. “Language” is a formal system of differences and by no means restricted to vocal utterances. In the language of speech, the spoken word “water” is not the spoken word “ice” because they have different forms, to which we attach different meanings. In the language of writing, Egyptian is not , though, both transliterate as sny: the one means “two” and the other means “companion.” Different meanings accompany different forms. Similarly, in the language of writing [$] means something different from [%] because they have different forms to which we assign different significations. These signs belong to the language of writing, and they refer to words, though they do not have phonetic value.
The broad category of “language” will also include Morse Code, semaphore, and American sign language, which may refer to speech, but can never be confused with it. Such forms of language as Unicode or mathematical notation do not refer to speech at all. In the study of writing we speak of the “underlying language” essential to deciphering an unknown script, so that we easily forget that writing is itself a kind of language. The confusion between “writing” and “language” is profound, ubiquitous, and disruptive, so that in a popular view Chinese “language” is the same as Chinese “writing,” a confusion that turns out, oddly, to be true once we understand how little Chinese writing has to do with Chinese speech.
Lexigraphic writing is based in speech, yes, but because we know of ancient speech only through written documents, it is easy to think that we are talking about “language” or “speech” when really we are talking about graphic representations that make use of spoken lexical elements, which may constitute in themselves a kind of language, but by no means intend to preserve actual speech. The intention is to communicate information, and for this purpose a graphic system with systematic phonetic ties to speech is a tool of earth-shaking power. It is not, however, a tool for the preservation of ancient speech.
For this reason the Sumerian speech was of use to nonSumerian Semitic scribes, because the relationships between graphic symbols and symbols originating in speech had been established by ancient usage. Just so, it was logical and practical for medieval Europeans to use Latin as a basic system for understanding across the polyglot confusion of mutually unintelligible local dialects and languages. Sumerian written in cuneiform (“wedge-shaped”) writing was a traditional system of signs for communicating information whether you were Sumerian or not, and as such worked well. For the same reason, during the dominance of Assyria over the Near East during the ninth to seventh centuries BC and of Persia in the sixth to fifth centuries BC, the West Semitic Aramaic script encoding lexical forms from the “Aramaic language” was used by nonAramaic-speaking scribes over an area stretching from the Mediterranean to northwest India.
For example, in Figure 0.1, from the palace at Nineveh of Tiglath Pileser III (ruled 745–727 BC), one of the most successful commanders who ever lived, a beardless eunuch on the left calls out a list of booty while the presumably Assyrian-speaking eunuch in the middle records the inventory in the contemporary Assyrian dialect by impressing cuneiform characters with a stylus into a waxed wood tablet. The presumably Assyrian-speaking eunuch scribe on the right makes a duplicate record (to prevent cheating?) by writing on a roll of papyrus or leather, certainly West Semitic Aramaic characters tied to Aramaic speech. The difference in writing medium, part of any writing tradition, accompanies a difference in script and “underlying language.”
Lexigraphic writing may refer to elements of speech, but in real speech we find extraordinary local and social differences, so great among English speakers that TV interviews in a regional English are often given subtitles for the greater English-speaking audience; that is, by means of alphabetic writing the speech is reduced to a standard form. Even in the same town speakers may not understand one another across differences in class and social background, although they “speak the same language.” In my own experience, once, in Alexandria, Egypt, my middle-class guide was unable to communicate with a street sweeper who may have known the location of our hotel, yet both spoke “Arabic.”
Only writing, and especially alphabetic writing supported by political power and social prestige, creates the illusion that a “language” such as English is a single thing, out there, bounded, defined, and capable of discovery. Writing’s overarching power stabilizes speech, represses local differences, and fashions standards for thought and expression. Dante’s Florentine dialect was one of hundreds of local Italic vernaculars descended from Latin, but his written Commedia in a fourteenth-century Florentine dialect created “the Italian language.” Many books speak of the Phoenician, Hebrew, Moabite, Aramaic, and Syriac “languages” when, really, they are looking at small variations in the forms of West Semitic writing applied over a broad geographical area to a single speech-family that we might loosely call “Semitic,” with local differences based on a similar phonology (a selection of voice sounds) and a similar inner structure. For example, in Hebrew the word for son is bn and in Aramaic it is br, but in Cambridge, Massachusetts, spirits to drink are /lika/ and in Seattle they are /likor/. Still, spoken communication takes place. The imperative “carry!” is qabur in the Hebrew “language,” qabor in Syriac, ‘uqbur in Arabic and qabar in Ethiopic, but all are mildly different expressions from a single underlying system. The great family of Semitic languages has very many regional variations, and we are simply never sure when a dialect has slipped over into a new “language,” that is, when a speaker within one system can no longer understand a speaker from another system. But twenty years of study of the holy Quran, certainly written in Arabic, will not enable the student to converse, even about simple things, with an inhabitant of Fez, Cairo, or Damascus, where everyone speaks “Arabic.”
The confusion is clear in Figure 0.2, a type of chart that appears in many books on writing. The chart catalogues the transformations undergone by the West Semitic signary (in which a hypothetical but wholly unproven priority is given to Phoenician script). Such graphic variations are taken as designating the different languages of “Phoenician,” “Moabite,” various forms of “Aramaic,” and “Hebrew.” But such different “languages” are as close to one another as Quranic Arabic is to spoken Arabic in its myriad and often mutually unintelligible varieties. It is true that, schooled in the Phoenician script of 1000 BC, one will have a hard time reading “Palmyrene Aramaic” of the third century AD, the script and language used in the caravan city of Palmyra in the Syrian desert, but these are nonetheless the same script with formal differences growing over more than one thousand years. The underlying “language” has remained the same. Such charts are really a study in handwriting, or paleography, with limited importance for understanding the theory and history of writing, and they do not describe an evolution of “language.”
In sum, the ambiguous correspondence of language and speech afflicts all such studies as this one, because speech, sometimes called “spoken language,” certainly is a language, but not all language is speech. We will need constantly to speak of the “language” underlying systems of writing, even if we really mean “speech.” In this case by “language” we refer to a system of phonic symbols intelligible to speaker and listener, more or less, over a wide range of variation. However, that system of phonic symbols could never itself be the “language” of the lexigraphic writing, because lexigraphic writing is its own language, making use of the resources of speech but never identical with it. “To speak a language” means something – if you speak Greek, you may enter the Mysteries at Eleusis; if you don’t, you can’t. Nonetheless, language, speech, and lexigraphic writing are all tangled up and, once we acknowledge their boundaries and differences, we must to some extent live with the confusion.
Transliterations
In the case of writing that is related to speech we must constantly deal with reconstructions of the sounds of speech encoded in the writing. In talking about the forms of characters and the sounds of speech I will follow the conventional practice of using brackets [] to enclose a given form or shape, how something looks, how it is written, and slashes / / to indicate the sound.
For example, in Egyptian hieroglyphs the sign represents the sound /m/ (plus an implied vowel). Unfortunately, each discipline has its own system, or more than one, for transcribing sounds, which originated in the history of the discipline. Such systems, internal to a discipline, are not themselves consistent, but may differ in England and Germany and even in the same place. For example, the Egyptian character for /y/ is written sometimes as [i] and sometimes as [j]. The situation is worse between disciplines. The glottal stop, when the throat closes as before and between “uh-oh,” is represented as [‘] in transliterations from West Semitic, but as [] by Egyptologists, although the sound is the same. Ideally, everyone might use the admirable International Phonetic Alphabet (IPA) as a means to suggest the sound values in systems of writing (see Fig. 14.3), and sometimes you do find this. However, few readers without formal training in linguistics understand the symbols of the IPA, the use of which can become a kind of game of phonetic precision that misrepresents the enormous phonetic ambiguity in all writing systems. Furthermore, the IPA is not easy to learn or to understand.
The problem of consistency in transliteration is not solvable, and in this book I have chosen, as much as possible, to use systems of transliteration traditional within a discipline, explaining myself as I go, but cautioning the reader about the need for flexibility and attention to sometimes subtle distinctions.