On the Theory of Writing

Abstract

MY FRIEND and colleague, Professor I. J. Gelb, has sent me his eagerly awaited book, A Study of Writing: The Foundations of Grammatology (London and Chicago, 1952). I welcome this book because of my deep conviction truth is to be discovered (if at all) only through the full and dispassionate discussion of the different opinions held by reasonable men, no matter how improbable this or opinion may seem to other reasonable Gelb and I have discussed aspects of writing face to face on numerous occasions in the past. On some points we agree; on others we differ. Several times in the present book Gelb has cited my writings among those with which he disagrees. I welcome this frank and direct approach to our divergencies of opinion, and my chief purpose here is to answer in the same spirit. Furthermore, Gelb's basic purpose-to lay the foundations of a science of writing -is one with which I wholeheartedly sympathize. I still cling to the slightly archaic assumption phenomena must be governed by natural laws. It seems axiomatic to me the operation of natural laws often produces random patterns. Apparently Gelb disagrees, for he objects to my characterization of Egyptian writing as utterly unsystematic' on the ground all life is governed by rules and principles (p. 170). If he means life, or even human life, is governed by rational rules and principles, I disagree. It is one thing to believe a group of phenomena must be governed by natural laws and a different thing to believe the relevant laws have been or can soon be discovered. Gelb believes he has discovered a specific principle of universal validity, that in reaching its ultimate de velopment writing, whatever its forerunners may be, must pass through the stages of logography, syllabography, and alphabetography in this, and no other, order (p. 201). Let us see what evidence he has mustered in support of this hypothesis. Syllabic signs developed out of word signs in Sumerian cuneiform and its offshoots (pp. 61-72 and 120-22), in Hittite hieroglyphic (pp. 81-85), in Chinese (pp. 85-88) and its Japanese offshoot (pp. 15962), and perhaps in Cypriote (pp. 196 and 215-17). Granting the case of Cypriote, this makes a total of four independent (or possibly independent) ancient systems which may be said to fit the first part of Gelb's hypothesis. Since none of these systems ever produced an alphabet, we have, so far, no evidence for the second part of the hypothesis. In his chapter vii (pp. 206-11), Gelb discusses a number of systems introduced in modern times among primitive societies under the influence of white men. Four of these show a logographic stage followed by a syllabic stage-to wit, Cherokee, various writings used among the Alaska Eskimos, Vai, and Bamum. Gelb states some of the Alaska Eski-